#92344: Childhood Leukemias and Lymphomas

Burkitt lymphoma is the most prevalent type of non-Hodgkin lymphoma among children/adolescents, accounting for approximately 23% of pediatric non-Hodgkin lymphoma in the United States [1]. It most commonly develops in males 15 to 19 years of age [1]. It is more common in non-Hispanic White adolescents than in Hispanic White adolescents [17]. This type of lymphoma was first noted by Denis Burkitt, who was studying children in Africa [50]. Burkitt lymphoma is endemic in Africa, where it occurs at a very high rate [17,51].

There are clinicopathologic differences between the endemic and sporadic forms of the disease [50]. Burkitt lymphoma most commonly affects the jaw and maxilla of African children, whereas the abdomen is the most common site in children in the United States [51].

Diffuse large B-cell lymphoma is the second most common non-Hodgkin lymphoma, representing 10% to 20% of pediatric non-Hodgkin lymphoma. It occurs most frequently in adolescents 15 to 19 years of age [17]. This type of lymphoma differs from other types of non-Hodgkin lymphoma in that it grows more slowly and does not usually spread to the bone marrow, central nervous system (CNS), or skin.

Diffuse large B-cell lymphoma is further categorized as nonmediastinal or primary mediastinal. Nonmediastinal disease is more common; primary mediastinal occurs predominately in adolescents [52]. The gene expression profile of this type of non-Hodgkin lymphoma is distinct and suggests a close association with Hodgkin lymphoma [53,54].

Lymphoblastic lymphoma represents approximately 16% of pediatric non-Hodgkin lymphoma [17]. It develops most often in boys, especially in the 15 to 19 year age group [17]. The malignant lymphoblasts are the same as those involved with childhood ALL. If the lymphoblasts involve more than 25% of the bone marrow, the disease is reclassified as leukemia and is treated accordingly. Lymphoblastic lymphoma can arise from early T cells or B cells; precursor B-cell lymphoblastic lymphoma occurs less frequently, representing 2.5% of all non-Hodgkin lymphoma [55].

Anaplastic large-cell lymphoma accounts for approximately 10% of pediatric non-Hodgkin lymphoma, occurring most often among male adolescents [1,17]. This type of non-Hodgkin lymphoma varies substantially in its clinical presentation, with possible involvement of lymph nodes and extranodal sites, typically bone and skin.

The following factors have been found to have prognostic significance for children with ALL [76,77]:

Age at the time of diagnosis is a strong prognostic factor, and the variation in outcome reflects differences in the underlying biology of ALL in different age groups [78,79]. For example, there is a high percentage of favorable cytogenetic features, such as high hyperdiploidy and t(12;21), in children who are 1 to 9 years of age, and the disease-free survival is better for that age group than for infants and older children and adolescents [80,81,82,83]. In addition, the risk of treatment failure is high for infants with ALL, especially infants younger than 3 months of age [84,85,86].

The risk of treatment failure is also high for children with precursor B-cell ALL who have an increased WBC count (more than 50,000/mcL) at the time of diagnosis [77,83]. In turn, a high WBC count is associated with other high-risk prognostic factors, such as t(4;11) and t(9;22).

The CNS status at the time of diagnosis is described according to results of a nontraumatic diagnostic lumbar puncture, specifically, the number of WBCs and the presence or absence of blasts on cytospin of the CSF sample. CNS3, which is defined as a WBC count of ≥5/mcL and the presence of blasts on cytospin, is associated with a higher risk of treatment failure compared with patients who do not have CNS involvement at the time of diagnosis. CNS2 status (WBC count less than 5/mcL and the presence of blasts) may be associated with a higher risk of relapse, but this risk may depend on treatment [87,88,89]. With respect to the sex of the patient, the prognosis has been slightly better for girls than for boys [90,91,92,93]. Although testicular relapse is one reason for this difference, relapse in bone marrow and CNS is also more frequent among boys [90,91,92]. Male sex has not been a risk factor in studies in which the event-free survival rate is more than 80% [94,95]. Studies that have examined sex differential for etiologic clues have identified several sex-specific genetic markers for childhood ALL risk [96,97,98,99] .

The role of race/ethnicity as a prognostic variable is unclear. Early studies indicated better survival rates for White children than for Black children, and later studies demonstrated lower survival rates for Black, Hispanic, and American Indian/Alaska Native children compared with White and Asian/Pacific Islander children/adolescents [100]. However, studies done at St. Jude Children's Research Hospital have indicated no racial/ethnic differences in survival, leading researchers to believe that survival depended instead on access to the same treatment protocols [86,101]. Social, biologic, and pharmacokinetic factors should be explored in an effort to better understand racial/ethnic variations in outcome [102,103]. Clinical and public health strategies can help improve access to health care, clinical trial enrollment, treatment, and survivorship care for children with ALL [104].

With regard to cytogenetic findings, hyperdiploidy is associated with a highly favorable prognosis and is also associated with other favorable prognostic factors, including an age of 1 to 9 years, a low WBC count, and t(12;21) with the TEL-AML1 fusion gene [77,80,81,105,106,107,108]. Hypodiploidy has been associated with progressively worse outcomes as the number of chromosomes decreases, with the worst outcome associated with near haploidy (24 to 28 chromosomes) [76,77,106]. Extreme hypoploidy and presence of the Philadelphia chromosome has been an indicator of very poor prognosis in B-cell ALL [77,109].

Chromosome translocations may influence prognosis, but other prognostic factors and the type of treatment may determine the effect of a specific translocation. TEL-AML1 fusion has been associated with an excellent outcome [68,81,110]. In contrast, the Philadelphia chromosome is related to an unfavorable prognosis, especially when it occurs in children who have a high WBC count or in whom the response to initial therapy is slow [60,111]. The t(4;11) has also been related to a high risk of treatment failure, and the event-free and overall survival rates are lower for children with Down syndrome and ALL [112,113,114].

The first prognostic factors identified for AML were age, WBC count, and FAB classification at the time of diagnosis, but their ability to predict outcome has been inconsistent [71,72]. A report of 61 children treated between 1976 and 1984 indicated that an age of younger than 2 years at the time of diagnosis, a WBC count of less than 100,000/mcL, and an FAB subtype of M4 or M5 were predictive of an increased risk of relapse and a decreased overall survival [72]. Adolescent patients are at higher risk for treatment-related mortality compared with younger children [115].

In another study of nearly 500 patients treated between 1986 and 1989, univariate and multivariate analyses demonstrated several clinical, morphologic, and cytogenetic factors associated with a lower rate of complete remission, including hepatomegaly, low platelet count (20,000/mcL or less), FAB subtype M5, and trisomy 8 [71]. In addition, a higher platelet count (more than 20,000/mcL) and absence of hepatomegaly were independent prognostic factors of better overall survival [71].

Data for a series of nearly 300 patients treated at St. Jude Children's Research Hospital from 1980 to 1996 indicated that the five-year event-free survival was slightly better for an age of 2 years or younger, an initial WBC count of less than 50,000/mcL, and an FAB subtype of M3 or M5 [22]. Although sex has not been used as a prognostic factor, several studies have suggested that the outcome is more favorable for girls than boys [22,71].

The ability of cytogenetic findings to predict outcome has varied, but they appear to be the best prognostic factors in AML [70,116,117]. In one study, trisomy 8 was the strongest predictor of poor prognosis, and in another, abnormalities of 11q23, q3, or chromosomes 5 or 7 were associated with a high rate of treatment failure [82]. In the series of patients treated at St. Jude, the best outcome was associated with t(9;21), and a favorable outcome has also been associated with t(8;21), t(15;17), and inv(16) [22,70]. Aberrations of chromosome 16q have also been shown to be a prognostic indicator of complete remission and better overall survival [69,71]. For children with monosomy 7, AML progresses rapidly and the response to treatment is poor [60,77]. In contrast to the situation in ALL, children with Down syndrome and AML have a good response to treatment [71,116]. Advances in technology are allowing researchers to identify other factors associated with outcome, and further studies are needed to confirm the usefulness of these factors as prognostic variables [118].

Children and adolescents with Hodgkin lymphoma are often asymptomatic, and the disease is frequently an incidental finding during a routine examination or evaluation of an injury. Approximately 25% to 30% of children will have one of three archetypal constitutional symptoms, known as B symptoms: unexplained, intermittent fever of 38 degrees Centigrade; drenching night sweats; or unintentional loss of more than 10% of body weight within the past six months (Table 8) [119]. The suspicion of Hodgkin lymphoma is increased when B symptoms are present with lymphadenopathy or splenomegaly. Patients may also have nonspecific systemic symptoms, such as fatigue and anorexia. In some patients, generalized pruritus may be present for months before lymphadenopathy is found, and excoriations may be evident as a result of excessive scratching [119]. On physical examination, the most common finding is a persistent, painless adenopathy, usually in the supraclavicular or cervical area [63,119]. As noted, reactive lymphadenopathy is common in children/adolescents, making it necessary to carefully evaluate patients to first rule out other infectious or inflammatory conditions. Involvement of the supraclavicular nodes should prompt earlier evaluation for Hodgkin lymphoma, as the cervical nodes are most commonly involved in infection and inflammatory conditions. Infectious mononucleosis can be distinguished from Hodgkin lymphoma by the presence of symmetrical cervical lymphadenopathy with pharyngitis. A tuberculin test may be helpful to rule out tuberculosis, which also may have similar clinical characteristics.

Involvement of only lymph nodes and/or the spleen is present in most patients [119]. In general, enlarged lymph nodes are firm and nontender. In advanced disease, enlarged nodes may grow together to form aggregate masses that become attached to underlying tissues [119]. Approximately 15% to 20% of patients will have noncontiguous extranodal involvement, and the lung, liver, bone marrow, and bone are the most common sites of extranodal involvement [123,124]. Extranodal involvement is more commonly associated with the mixed cellularity subtype of Hodgkin lymphoma, whereas involvement of the cervical, supraclavicular, and mediastinal lymph nodes is most often associated with the nodular sclerosis subtype [119].

The mediastinum is involved in about two-thirds of patients [119]. Symptoms accompanying mediastinal involvement may include a persistent, nonproductive cough, dyspnea, or chest pain. Evaluation for Hodgkin lymphoma should be carried out if these symptoms are present, especially if accompanied by B symptoms. It is important to distinguish involvement of the mediastinum from a large, normal thymus or from other mediastinal tumors [27,119]. Morphology, clinical signs, radiographic findings, and immunophenotyping findings must all be evaluated closely to make the distinction.

Non-Hodgkin lymphoma should be considered in the differential diagnosis of Hodgkin lymphoma. One distinguishing feature is the rate of disease progression; involved lymph nodes will grow more rapidly in non-Hodgkin lymphoma than in Hodgkin lymphoma. In addition, abnormal laboratory values are more common with non-Hodgkin lymphoma [119].

The diagnostic work-up for suspected Hodgkin lymphoma includes laboratory testing and imaging studies to evaluate the site and extent of disease. Laboratory testing should begin with a CBC with differential and an erythrocyte sedimentation rate. The results of the CBC may be nonspecific, demonstrating neutrophilia, eosinophilia, and thrombocytosis [27,119]. Lymphopenia and normochromic normocytic anemia are indicators of extensive disease [125,126]. In the presence of other signs of Hodgkin lymphoma, elevations of the erythrocyte sedimentation rate, lactate dehydrogenase level, and ferritin level may be further indicators of Hodgkin lymphoma [63]. An elevated level of alkaline phosphatase may indicate metastatic bone disease, and further testing should be done to determine if there is skeletal involvement [119].

Evaluation of a bone marrow sample is usually not routinely done. Bone marrow involvement is a characteristic of primarily lymphocyte-depleted Hodgkin lymphoma, which is rare in the pediatric population; approximately 10% to 15% of children/adolescents will have bone marrow involvement at the time of the initial diagnosis [27]. The frequency of bone marrow involvement is higher for older male children who have constitutional symptoms. Bone marrow biopsy should be limited to patients who have B symptoms and in whom advanced disease is suspected [119].

A diagnosis of Hodgkin lymphoma can be confirmed only through evaluation of a specimen from an involved lymph node, and biopsy should be done when there is no response to a course of antibiotics. The biopsy technique depends on the site of the involved node. Excisional biopsy is often the preferred technique, as it allows for better determination of the histologic subtype [119]. However, the least invasive procedure should be used [127,128]. Thus, fine-needle aspiration may be more appropriate if the involved node is in the thoracic or abdominal cavity, and CT can be used to guide biopsy in these situations.

As noted, Hodgkin lymphoma is classified histologically according to four subtypes. Nodular sclerosis is the most common subtype, representing approximately 72% of the cases of pediatric Hodgkin lymphoma. This subtype occurs most frequently in adolescents, and the mediastinum is involved in 80% of cases [27]. Mixed cellularity Hodgkin lymphoma accounts for approximately 25% of pediatric Hodgkin lymphoma, most frequently developing in children younger than 10 years of age [27]. It is usually associated with the peripheral lymph nodes of the upper part of the body.

The American Joint Committee on Cancer has adopted the Lugano classification system to evaluate and stage lymphoma. The Lugano system replaces the Ann Arbor classification system (which was adopted in 1971). The Lugano system categorizes the disease as stage I through IV according to the extent of lymph node involvement (Table 9) [129]. The system is based on the premise that Hodgkin lymphoma progresses along contiguous lymph nodes [130]. Within this classification system there are additional designations that are applied to further define symptoms and disease [130].

According to the National Comprehensive Cancer Network, there is no uniform risk stratification for pediatric Hodgkin lymphoma. However, several factors are considered to confer poor prognosis, including B symptoms, extranodal disease, Ann Arbor stage, gender, and response to initial chemotherapy.

(https://www.nccn.org/professionals/physician_gls/pdf/ped_hodgkin.pdf Last Accessed: August 22, 2024)

Level of Evidence: Expert Opinion/Consensus Statement

Prognostic variables have less importance in Hodgkin lymphoma because of advances in treatment, but some variables are helpful in predicting outcome. The most significant prognostic indicator is stage of disease. The outcome associated with stage I, II, or III disease is better than that for stage IV disease [47,123,131]. Studies in which prognostic factors were evaluated in multivariate analysis have demonstrated the following factors to be of import [131,132]:

The presence of B symptoms (low-grade fever, night sweats, and weight loss) reflects aggressive disease and has been associated with a worse prognosis [131,132,133]. High tumor burden is also predictive of a worse prognosis; a ratio of mediastinal mass-to-thoracic cavity of more than one-third has been associated with a greater risk of recurrence and with a slightly lower survival rate in several studies [131,133,134]. The number of disease sites may also play a role in predicting outcome. Patients who have fewer than four sites of involvement generally have a better outcome [129,135].

Several laboratory studies, including the WBC count, hemoglobin, erythrocyte sedimentation rate, and serum ferritin level, have also been correlated with a poor prognosis [129,131,136]. In one study, a WBC of more than 13,500/mcL and a hemoglobin level of less than 11.0 g/dL were significantly associated with shorter disease-free survival [131].

The predictive value of patient-related factors such as sex and age is unclear. One study indicated that disease-free survival was better for girls than for boys [131]. Age has been an important indicator in some studies, with 5-year and 10-year survival rates being higher for children 10 years of age or younger than for older children and adolescents (11 to 16 years of age) or young adults [137].

The rate of response to initial cycles of chemotherapy appears to be prognostically important and is being used in the research setting to tailor subsequent therapy, avoid treatment-associated risks that compromise long-term health, and achieve high cure rates [47,138].

The clinical evaluation and diagnostic testing for non-Hodgkin lymphoma is similar to that for Hodgkin lymphoma. Lymphadenopathy should be evaluated, and infectious or inflammatory conditions should be ruled out before a biopsy is performed. The sites of disease and associated symptoms vary according to the histopathologic type of non-Hodgkin lymphoma [120,121,122]. Most children/adolescents will have advanced disease at the time of presentation [17,121].

B-Cell Lymphoma (Burkitt, Burkitt-like, and Diffuse Large B-Cell)

Diffuse large B-cell lymphoma and Burkitt or Burkitt-like lymphoma may appear clinically similar, but the former is more often localized [139,140]. The most common site of B-cell lymphoma is the abdomen, and involvement of peripheral lymph nodes is relatively uncommon [17,120]. When present, peripheral lymph node involvement is more likely to be associated with diffuse large B-cell lymphoma, with sites in the neck, mediastinum, or throat [120]. Symptoms associated with B-cell lymphoma in the abdomen include abdominal pain, nausea, vomiting, and diarrhea. A mass may be palpable on physical examination of the abdomen [63,120]. Persistent abdominal pain with vomiting or diarrhea, especially when accompanied by significant weight loss, should prompt further diagnostic testing. Intussusception of the small bowel is common.

Lymphoblastic Lymphoma

In comparison with B-cell lymphoma, the abdomen is rarely involved with lymphoblastic non-Hodgkin lymphoma. Instead, the mediastinum is involved in nearly 75% of children/adolescents with T-cell lymphoblastic lymphoma, as a result of development of the lymphoma from the thymus [17]. The mass forms in the anterior mediastinum. Dyspnea is often the first symptom [121]. Coughing, wheezing, stridor, dysphagia, and superior vena cava syndrome may also be presenting symptoms. Because mediastinal involvement with non-Hodgkin lymphoma progresses more rapidly than with Hodgkin lymphoma, these symptoms may be more severe. Pleural effusion may also occur [17]. Localized lymphoblastic lymphoma may occur in lymph nodes, bone, and subcutaneous tissue. Other lymph nodes, such as the tonsils or neck nodes, are less commonly the primary site of T-cell lymphoblastic lymphoma. Disease spread can be rapid, with the brain, pleura, pericardium, and bone marrow being the most common sites of metastases. Disease is in an advanced stage (stage III or IV) at the time of diagnosis in most cases [121].

In the case of precursor B-cell lymphoblastic lymphoma, disease is usually more limited and involves skin, soft tissue, bone, tonsil, or a sole peripheral nodal region [121].

Anaplastic Large-Cell Lymphoma

The course of symptoms with anaplastic large-cell lymphoma may be indolent, with fever and a waxing and waning of lymphadenopathy [141,142]. Nearly 50% of patients have B symptoms, and 88% to 97% of patients have lymphadenopathy, with the peripheral nodes being involved most often. CNS and bone marrow involvement occur less frequently with anaplastic large-cell lymphoma than with other types of non-Hodgkin lymphoma [17,122,142].

Soft tissue and skin are the most frequent sites of extranodal involvement [17,122,142,143,144]. Cutaneous manifestations may be single or multiple nodules, multiple or disseminated red-yellow papillomatous lesions, or large ulcerated lesions [122]. Other extranodal sites include the gastrointestinal tract, lung, pleura, and muscle.

As with Hodgkin lymphoma, the diagnostic work-up for suspected non-Hodgkin lymphoma includes laboratory testing, diagnostic imaging, and biopsy of suspicious nodes. A CBC can help distinguish non-Hodgkin lymphoma from infectious or inflammatory adenopathy [121]. A high level of serum lactate dehydrogenase level is often indicative of a large mass, but the level is increased less often in patients with anaplastic large-cell lymphoma than in patients with B-cell lymphoma or lymphoblastic lymphoma [122]. A CSF sample should be examined, and evaluation of biopsy specimens should include histopathologic analysis, immunophenotyping, and cytogenetic and molecular analysis [120,121,122,142].

The use of imaging studies is similar for all types of suspected lymphoma. The choice of study is based on the suspected site of involvement. A chest x-ray or CT of the chest is indicated if a mediastinal mass is suspected, which is often associated with T-cell lymphoblastic lymphoma or diffuse large B-cell lymphoma [120]. MRI of the head and neck should be done if signs or symptoms are suggestive of involvement of neck nodes. The abdomen is the most frequent site of involvement for B-cell lymphoma, making CT of the abdomen perhaps the most valuable imaging study in that setting. CT also is useful for staging the lymphoma and for differentiating the primary cutaneous form from the systemic form [120,145]. Skeletal scintigraphy is useful for detecting bone involvement, such as with anaplastic large-cell lymphoma [142]. Positron emission tomography (PET)/CT scanning has largely supplanted scintigraphy and is highly recommended for patients with suspected non-Hodgkin lymphoma [146,147,148]. PET/CT scanning is also useful for assessing response to treatment and for identifying occult sites of disease. Whole-body scanning may be helpful in detecting disease sites that would otherwise be unidentified [120,145]. If advanced disease is suspected, MRI or CT of the brain should be done [145].

Pediatric non-Hodgkin lymphoma is classified as stage I through IV according to the St. Jude staging system, which is similar to the Ann Arbor staging system for Hodgkin lymphoma [149,150]. These stages are [17,149,150]:

As with Hodgkin lymphoma, advances in treatment for non-Hodgkin lymphoma have made prognostic variables less significant [120]. In addition, factors predictive of outcome vary according to the histologic subtype of non-Hodgkin lymphoma.

For B-cell lymphomas, the most important prognostic factor is tumor burden, as determined by either stage of disease or the serum lactate dehydrogenase level [120]. The tumor burden is high for patients with extensive bone marrow involvement (more than 25%) [120]. Serum lactate dehydrogenase level is a well-recognized risk factor that correlates with tumor burden, and a level of more than 1,000 IU/L has been associated with a poor prognosis [52,151,152].

Age and sex have also been evaluated as prognostic factors in B-cell lymphoma; many studies have not found them to be predictive of outcome [120]. In one study involving children/adolescents with diffuse large B-cell lymphoma, the event-free survival was significantly better for male patients than for female patients and the outcome was better for younger girls than for adolescent girls [139]. Again, the universal applicability of these findings is not established.

In an analysis of pooled data from three multicenter studies of prognostic variables in anaplastic large-cell lymphoma, multivariate analysis demonstrated that the following factors were significantly associated with a risk for treatment failure [120,144,152,153,154,155]:

There was no significant effect of age, sex, or stage of disease in any of the studies [17].

In a comprehensive analysis of potential prognostic factors for outcome in lymphoblastic lymphoma, no factor was found to be associated with outcome [156].

A cancer diagnosis presents the dual challenge of being both emotionally and intellectually overwhelming. These feelings are heightened for parents of a child/adolescent with cancer. Parents differ in how much information they want or feel they can handle about the diagnosis. Clinicians should ask how much detail the parents wish to hear [181]. Important points for communicating the diagnosis include:

Clinicians should address parents' and patients' fears and reassure them that they will be available throughout the course of treatment [181]. Eliciting and validating emotional reactions at this time will help parents and patients better understand and comply with further diagnostic testing [181].

Guidelines are available to help clinicians deliver bad news in the childhood cancer setting with sensitivity (Table 11) [184,185,186]. However, the stress of receiving bad news, even if the clinician is compassionate and speaks clearly, can prevent patients and family members from retaining needed information [185]. After relating bad news, clinicians can build a partnership by allowing parents to express their feelings and opinions [187]. A pediatric psychologist or family therapist can assist both the child and family in coping with the impact of the diagnosis.

Clinicians should also determine parents' preference for the type and amount of information to be shared with the child. Many parents wish to protect their child by withholding information. This is especially true among some cultures—for example, among Chinese, Japanese, and Greek populations [188]. However, studies have shown that children often recognize the seriousness of their illness and prefer open communication about their disease and prognosis [188,189,190]. Such open exchange of information can help to avoid the fear of the unknown and to preserve the child's trust in his or her parents and/or family and caregivers [188,190]. Even if the child is to be included in the discussion, the best approach may be to sequence some information; that is, to communicate information to parents in separate meetings before communicating it to the child. Parents of children with leukemia have acknowledged the benefits of communicating openly with children, but they noted that their child's presence during discussions of such topics as prognosis, treatment options, and adverse events, restricted their own communication with physicians, made it difficult to concentrate, and interfered with their efforts to care for their child emotionally [191]. Parents also said that separate meetings allowed them to absorb information and to convey it to their child at an appropriate time and in a reassuring way [191].

Parents and clinicians should involve the child in the discussion of the diagnosis using language that is developmentally appropriate for the child, checking often to make sure the child understands. Having the child repeat the information in his or her own words is one way to assess comprehension [188]. The child should also be encouraged to express feelings and concerns. Adolescents and children who are mature enough to understand and reason should be allowed to express their opinions, and healthcare professionals should be advocates for the child's preferences and decisions [183]. Adolescents especially, because of their need for independence, should be given choices about their care, whenever possible, as well as about privacy issues, such as with whom information can be shared [8].

Clinicians should explore ways to best communicate with and educate children/adolescents. Computer-assisted and multimedia educational programs have been shown to enhance cancer-related knowledge among children/adolescents and improve their feelings of control over their health [192].

Risk-adapted treatment protocols developed in the 1990s have achieved a cure rate of 80% [101,224]. Most pediatric study groups classify patients with ALL according to three risk categories: standard, high, and very high; COG uses a fourth category for very low probability of relapse [223,225].

In general, treatment of pediatric T-cell or precursor B-cell ALL consists of intensive multidrug chemotherapy for standard-risk patients, allogeneic HSCT for high-risk patients, and antimetabolites for low-risk or very-high-risk patients [41]. All children with ALL receive CNS prophylaxis, primarily with intrathecal chemotherapy [74,220]. Prophylaxis with cranial radiation is reserved for patients at very high risk of relapse, such as patients with CNS involvement at diagnosis or with T-cell ALL [74,220].

Remission Induction

The goals of remission induction therapy are to induce complete remission (defined as eradication of 99% of the initial burden of leukemia cells), to restore normal hematopoiesis, and to regain a normal performance status [41,220]. This phase of treatment typically lasts four weeks [220]. The rate of complete remission after induction therapy is high, ranging from 96% to 99% [74,220]. Of the patients in whom complete remission is not achieved within the first four weeks of treatment, approximately 50% will experience a toxic death during induction (usually due to infection) and the remaining 50% will have resistant disease [94,156,220,226,227]. The protocols for children with high-risk or very-high-risk ALL and all adolescents with ALL involve treatment with at least four chemotherapeutic agents. These agents most often include a corticosteroid (prednisone or dexamethasone), vincristine, and asparaginase and/or an anthracycline (doxorubicin or daunorubicin) [220,228]. The COG protocols reserve the use of a four-drug induction for patients with high-risk B-cell ALL and T-cell ALL and do not include administration of anthracycline during induction to patients with standard-risk ALL [220,228]. Patients treated by three other study groups receive an induction regimen with four or more drugs regardless of presenting features [220]. A meta-analysis of studies comparing the use of dexamethasone or prednisone showed that dexamethasone was more efficacious, with lower rates of events (death from any cause, refractory or relapsed leukemia, second malignancy, and CNS relapse) but no difference in bone marrow relapse or overall mortality [229]. Dexamethasone was associated with higher rates of toxicity, however, and it is not clear whether the short-term benefits found will translate to better overall survival [229]. A randomized trial comparing doxorubicin and daunorubicin given during induction found no difference in efficacy between the agents in early response measures [230]. In June 2021, the U.S. Food and Drug Administration (FDA) approved Rylaze (Erwinia asparaginase [recombinant]) as a component of the multi-agent chemotherapy regimen to treat ALL (and lymphoblastic lymphoma) in adult and pediatric patients (one month and older) who have developed hypersensitivity to Escheria coli-derived asparaginase products [231]. Rylaze was found to maintain a clinically meaningful level of asparaginase activity throughout the entire duration of treatment [231]. In 2022, the FDA approved a new dosing regimen for Rylaze in adult and pediatric patients with ALL who are 1 month of age or older and who have developed hypersensitivity to E. coli-derived asparaginase [232].

The emergence of targeted therapies for treatment of Philadelphia chrome-positive (Ph+) disorders with tyrosine kinase inhibitors (TKIs) represents an important advancement in ALL therapy. Imatinib mesylate and dasatinib are inhibitors of BCR-ABL tyrosine kinase and approved by the FDA for treatment-naïve patients with pediatric Ph+ ALL [228]. In one study of early response to TKIs during remission induction in children with Ph+ ALL, TKIs (imatinib and dasatinib) produced a marked drop in minimal residual disease levels. At the end of remission induction, 9 of 11 patients who received imatinib or dasatinib and conventional induction chemotherapy had remission, compared with two of 16 patients who received chemotherapy alone. The five-year event-free survival rate was 68.6% for the 11 patients who received TKIs compared with 31.6% for the 19 patients who did not [233].

Consolidation/Intensification Therapy

Consolidation/intensification therapy begins after normal hematopoiesis has been restored. The most commonly used intensification regimen was first introduced by the Berlin-Frankfurt-Munster (BFM) clinical trials group and usually includes methotrexate with 6-mercaptopurine (6-MP) and high-dose asparaginase, followed by reinduction treatment (the same regimen given during the first few months of remission induction therapy). Maintenance typically includes 6-MP, weekly low-dose methotrexate, and sometimes administration of vincristine and a corticosteroid, as well as continued intrathecal therapy [74,220,228]. This regimen, with variations (e.g., intensification for higher-risk patients, use of escalating doses of methotrexate without leucovorin rescue, elimination or truncation of some phases for lower-risk patients), has been adopted by COG and has been associated with favorable outcomes in children with standard-risk ALL [234,235].

The doses of drugs are increased according to risk, and studies have been done to evaluate the efficacy of higher doses in preventing relapse. Research reported in 2011 demonstrated a significant improvement in event-free survival for children/adolescents with high-risk B-cell ALL when high-dose methotrexate without asparaginase was used compared with standard-dose methotrexate and asparaginase (82% vs. 75%) [236]. Similarly, compared with standard-dose methotrexate, high-dose methotrexate led to significantly better 5-year and 10-year event-free survival for children/adolescents with high-risk T-cell ALL, with rates of 80% vs. 68% and 77% vs. 66%, respectively [237].

In a landmark multicenter COG study of 92 children (1 to 21 years of age) with Ph+ ALL, investigators evaluated whether imatinib (340 mg/m² per day) with an intensive postinduction chemotherapy regimen improved outcome. Continuous imatinib exposure improved outcome in patients treated with imatinib for 280 continuous days, with a three-year event-free survival of approximately 80%, more than twice that of historical controls. Three-year event-free survival was similar for patients treated with chemotherapy plus imatinib or sibling donor blood and marrow transplantation. There were no significant toxicities associated with adding imatinib to intensive chemotherapy [238]. A follow-up study found that five-year disease-free survival was similar for patients treated with chemotherapy plus imatinib (70%), sibling donor blood and marrow transplantation (65%), and unrelated donor blood and marrow transplantation (59%). The reinduction rate following relapse was similar to other higher-risk ALL groups [239].

Another study assessed the safety and efficacy of oral imatinib in association with a Berlin-Frankfurt-Munster intensive chemotherapy regimen and allogeneic stem-cell transplantation for patients (1 to 18 years of age) with pediatric Ph+ ALL [240]. Patients were enrolled by 10 study groups between 2004 and 2009, and were classified as either good risk or poor risk according to early response to induction treatment. Good-risk patients were randomly assigned to receive postinduction imatinib with chemotherapy or chemotherapy alone. All poor-risk patients received postinduction imatinib with chemotherapy. All patients received four postinduction blocks of chemotherapy, after which they became eligible for stem-cell transplantation. Four-year disease-free survival was 72.9% in the good-risk/imatinib group compared with 61.7% in the good-risk/no imatinib group. Four-year event-free survival for poor-risk patients was 53.5% [240].

Continuation (Maintenance) Therapy

Continuation therapy usually consists of weekly methotrexate and daily 6-MP for a period of 2 to 3 years [74,220,228]. A review of trials involving more than 10,000 children demonstrated that shorter durations of treatment were associated with a higher risk of relapse and death [241,242].

CNS Prophylaxis

Before the routine use of CNS prophylaxis, relapse involving the CNS occurred in approximately 80% of all patients with complete remission [243]. Preventing relapse of disease in the CNS has led to significant improvement in the rate of isolated CNS relapse, which now ranges from less than 2% to 10% [22,101]. Radiation was first used for CNS prophylaxis, but this approach was associated with significant acute and late morbidity, including neurocognitive impairment, second cancers, and endocrinopathy [74,215]. As such, the use of intrathecal chemotherapy, usually methotrexate or cytarabine, is now used in most cases [220]. Children/adolescents with very-high-risk disease may still benefit from cranial radiation, and most treatment protocols limit the use of radiation to these instances, with the use of lower doses [156,220,244].

Treatment for Relapsed or Refractory Disease

Relapse occurs in approximately 25% of patients with ALL [245]. The most common sites of relapse in ALL are bone marrow, the CNS, and the testicles [245]. For bone marrow relapse, reinduction therapy (usually a four-week course of prednisone, vincristine, and an anthracycline, with perhaps the addition of asparaginase) is followed by postremission chemotherapy [245]. Treatment of CNS relapse usually involves intensive systemic therapy, with cranial or craniospinal radiation delayed for six months [246].

In 2017, the FDA approved the first gene therapy for the treatment of B-cell precursor ALL in children and young adults [247]. Tisagenlecleucel is a genetically modified autologous T-cell immunotherapy used in the treatment of refractory or relapsed disease. With this approach, the patient's T cells are collected and modified to include a chimeric antigen receptor that directs T cells to kill leukemias cells, then infused back into the patient. In one study, the three-month remission rate was 83% [247]. However, tisagenlecleucel is associated with potentially severe side effects, including cytokine release syndrome, neurologic events, serious infections, acute kidney injury, and hypoxia [248].

Isolated testicular relapse usually occurs later than either bone marrow or CNS relapse and is treated with intensive chemotherapy that includes high-dose methotrexate. Patients who do not have a complete remission after induction also receive local radiation therapy [220]. Orchiectomy of the clinically involved testis is performed in some European clinical trial groups, with biopsy of the contralateral testis to determine whether additional local control (i.e., surgical removal or radiation) is needed. One study that examined testicular biopsy at the end of frontline therapy failed to demonstrate a survival benefit for patients with early detection of occult disease [249]. While there are limited clinical data concerning outcome without the use of radiation therapy or orchiectomy, the use of chemotherapy (e.g., high-dose methotrexate) to achieve antileukemic levels in the testes is being tested in clinical trials [220].

Hematopoietic Stem Cell Transplantation

In 2012, the American Society for Blood and Marrow Transplantation (ASBMT) published an update to its 2005 recommendation that HSCT be done after first complete remission of Ph+ ALL when a matched related donor is available [250,251]. Allogeneic HSCT is recommended for children who [251]:

Allogeneic HSCT is not recommended for patients with T-cell ALL in their first complete remission; patients with the MLL gene plus ALL when it is the sole adverse risk factor; and patients with isolated CNS relapse in B-cell ALL [251]. In general, HSCT approaches benefit only children at high risk for relapse with standard chemotherapy approaches [252].

Infection prophylaxis during HSCT includes confinement in a HEPA-filtered, positive-air-pressure sealed room, strict hand hygiene, antibacterial prophylaxis with fluoroquinolone, and antifungal prophylaxis with fluconazole [253].

Follow-Up and Surveillance

Several organizations have published consensus guidelines for follow-up for late effects following HSCT; however, these guidelines do not address all pediatric-specific challenges to HSCT [254]. The Pediatric Blood and Marrow Transplant Consortium (PBMTC) published six detailed papers to address the lack of detailed pediatric-specific late effects and guidelines for long-term follow-up after HSCT. The PBMTC summary and guideline recommendations provide the most current outline for following up children for late effects after HSCT (Table 14) [255].

As noted earlier, the event-free and overall survival rates are lower for children/adolescents with AML than for those with ALL. Due to lack of consistent prognostic factors in AML, risk-adapted treatment is not regularly used as it is for ALL; however, risk-adapted and biologically targeted therapies are being tested to improve treatment while sparing normal tissues [44,256]. Optimal treatment of AML requires control of bone marrow and systemic disease [45]. The four-phase approach is not always followed, as the efficacy of the continuation (maintenance) phase has not been proven for the treatment of AML [22,44,45]. The approach to the treatment of AML is similar for all types except for the M3 subtype (acute promyelocytic leukemia), for which chemotherapy is combined with all-trans retinoic acid (ATRA) [22].

The drugs used to treat AML have primarily remained the same, but regimens and timing of treatment have been refined, with current rates of 80% to 90% for complete remission, 30% to 40% for relapse, 45% to 55% for event-free survival, and 55% to 65% for overall survival [44,45].

Induction Therapy

The most effective chemotherapy regimen for induction in AML has been an anthracycline, cytarabine, and etoposide [44,45]. Although the regimens used have differed in many ways, including the cumulative doses of drugs, the choice of anthracycline, and the number of treatment cycles, the results have been relatively similar [44,45].

Consolidation/Intensification Therapy

Studies and one phase III clinical trial have demonstrated that intensive treatment during the consolidation phase leads to decreased relapse rates and better event-free and overall survival for younger patients with AML [22,257]. High-dose cytarabine, usually given with mitoxantrone, has led to improved outcomes for children with high-risk disease [45].

CNS Prophylaxis

Intrathecal chemotherapy is considered by most to be a standard component of treatment of AML, although no studies have compared CNS prophylaxis with no CNS prophylaxis, and it has not been shown to contribute directly to improved survival [22,45]. Cytarabine, methotrexate, or both, along with hydrocortisone, are the preferred agents for prophylaxis [45]. The rate of isolated CNS relapse is reported to be 2% [44]. Cranial radiation is not used for prophylaxis [44,45].

Treatment for Relapsed or Refractory Disease

Most relapses occur during the first year, and the most common site of relapse is the bone marrow (92%) [258]. Relapse occurs in the CNS in 10% to 20% of patients and in the skin in 4%.

Remission induction chemotherapy regimens have included high-dose cytarabine in combination with fludarabine or mitoxantrone [44,45]. Anthracyclines are effective, but the increased risk for cardiotoxicity limits their use in the relapse setting if they have been used for primary treatment [258]. Liposomal daunorubicin has been added to fludarabine plus cytarabine in an attempt to limit cardiotoxicity [44]. However, cure is unlikely with chemotherapy alone for relapsed AML [44]. In 2020, the FDA extended the indication for gemtuzumab ozogamicin to include newly diagnosed CD33-positive AML in pediatric patients 1 month of age and older [259]. The drug was previously approved for children 2 years of age and older with relapsed or refractory AML [45,260]. Efficacy and safety information for the extended indication was supported by data from a multicenter randomized study of 1,063 patients, 0 to 29 years of age, with newly-diagnosed AML. The estimated percentage of patients free of induction failure, relapse, or death at five years was 48% in the gemtuzumab ozogamicin plus chemotherapy arm versus 40% in the chemotherapy alone arm. No difference in overall survival was found between treatment arms. The most common adverse events included cardiotoxicity, hepatotoxicity, infection, fatigue, and fever [260]. In a report from the COG on the efficacy of clofarabine in combination with cytarabine in pediatric patients with recurrent AML, the combination yielded an acceptable response rate without excess toxicity. The nearly 50% survival rate reported suggests that this combination is an effective bridge to HSCT [261]. HSCT is thought to be the only curative treatment for relapse [44,45,257,258,262].

Stem Cell Transplantation

The issue of HSCT in AML, especially with regard to when it should be done, remains a topic of some debate. A meta-analysis demonstrated that transplantation reduced the risk of relapse and significantly improved disease-free and overall survival among many children/adolescents with AML [263]. The review of evidence by the ASBMT led to its recommendation of transplantation after either first or second remission if a matched related donor is available [264]. However, a later meta-analysis showed that transplantation after first complete remission did not lead to significant differences in relapse or disease-free or overall survival compared with no transplantation [265]. Current application of allogeneic HSCT involves incorporation of risk classification to determine whether transplantation should be pursued in first remission [45]. Due to improved outcomes in patients with favorable prognostic features and lack of demonstrated superiority of HSCT, this patient population typically receives matched-family donor HSCT only after first relapse and achievement of a second complete remission [264,266,267,268].

The absence of published recommendations specific for pediatric AML motivated an international group of pediatric hematologists and oncologists to develop evidence- and expert opinion-based consensus recommendations for the management of AML in children [65]:

Follow-Up and Surveillance

The improved outcome in children with AML over the last 10 years is probably associated with better supportive care strategies [65]. Children with AML who receive treatment with contemporary therapy and remain in remission four years from diagnosis probably are cured. Although late relapses and late deaths from other causes are rare, long-term follow-up of survivors is necessary for the timely management of late adverse effects [269].

The National Comprehensive Cancer Network recommends that treatment of patients with favorable-risk pediatric Hodgkin lymphoma typically involve participation in an ongoing clinical trial.

(https://www.nccn.org/professionals/physician_gls/pdf/ped_hodgkin.pdf Last Accessed: August 22, 2024)

Level of Evidence: Expert Opinion/Consensus Statement

Standard treatment options achieve cure in approximately 80% to 90% of children/adolescents with Hodgkin lymphoma. The rates have been better for children 10 years of age and younger than for children older than 10 years of age [119,270]. Between 2011 and 2017, the five-year overall survival rate was 98% for patients younger than 20 years of age [47]. The German Hodgkin Study Group found that adult treatment protocols were safe and effective treatment options for adolescents with Hodgkin lymphoma [271]. No single treatment approach is ideal for all pediatric and young adult patients due to differences in age-related developmental status and sex-related sensitivity to chemotherapy toxicity. The general treatment strategy for children and adolescents with Hodgkin lymphoma is chemotherapy, with or without radiation [47].

The risk-adaptive approach to the treatment of Hodgkin lymphoma involves planning treatment according to prognostic variables such as stage of disease, presence of B symptoms, and tumor bulk [272]. Disease is classified as being low, intermediate, or high risk. Low-risk disease is defined as localized Hodgkin lymphoma (stage I or II and sometimes IIIA) with no B symptoms or tumor bulk. Intermediate-risk disease usually includes stage I or II disease with unfavorable features (B symptoms, tumor bulk, involvement of three or more lymph node regions, and/or extranodal extension to contiguous structures) [119]. Stages III and IV typically represent high-risk disease.

Ongoing trials for patients with favorable-risk (low-risk) disease are evaluating the effectiveness of treatment with fewer cycles of combination chemotherapy alone that limit doses of anthracyclines, alkylating agents, and radiation therapy. Contemporary trials for patients with intermediate/unfavorable disease are testing whether chemotherapy and radiation therapy can be limited in patients who achieve a rapid early response to dose-intensive chemotherapy regimens. Trials are also testing the efficacy of regimens integrating novel, potentially less-toxic agents such as brentuximab vedotin [47].

Favorable-risk (low-risk) disease is defined differently by different clinical trial groups, but for the most part it encompasses patients with localized stage I and II disease without adverse prognostic features (e.g., "B" symptoms, extranodal extension, mediastinal bulky disease). Favorable-risk disease can be treated with reduced therapy, which consists of two to four cycles of multiagent chemotherapy and low-dose (15–25 Gy), involved-field radiation [47,272,273]. The regimens used most often include etoposide, which has commonly replaced alkylating agents and anthracyclines, in an effort to reduce gonadal toxicity and enhance antineoplastic activity [47]. Intermediate-risk disease is treated with three to six cycles of compacted, dose-intensive chemotherapy and low-dose, involved-field radiation, usually as consolidation therapy. For high-risk disease, four to six compacted, dose-intensive cycles of chemotherapy in addition to low-dose (15–25 Gy) involved-field radiation therapy to involved sites of disease are recommended [272]. The dose-intensive regimen of bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone led to response within four weeks in 74% of children/adolescents with stage IIB or IIIB Hodgkin lymphoma with bulk disease or stage IV disease [274]. When followed by guideline-directed consolidation therapy, the treatment led to overall survival of 97% and five-year event-free survival of 94% [274]. More recent pediatric trials conducted in North America and Europe have used procarbazine-free standard backbone regimens, which represent dose-dense therapies that use six drugs to maximize intensity without exceeding thresholds of toxicity [47]. Early response to therapy may be considered in determining the need for radiation in those who have complete remission. Advanced radiation techniques, such as intensity-modulated radiation therapy and proton therapy, may be considered depending on the clinical scenario and if an improvement is expected [272,275].

In addition to disease-related factors, the patient's age and sex are primary considerations. A younger age at the time of diagnosis has implications for musculoskeletal and soft-tissue deformities and cardiovascular dysfunction after radiation and/or chemotherapy with an anthracycline [119]. With respect to the patient's sex, the risk of breast cancer is increased for teenage girls who have radiation to the thorax, and girls are at higher risk of cardiomyopathy as a result of anthracycline use [119]. Alkylating agents lead to a higher prevalence of gonadal dysfunction in boys. The German Paediatric Oncology and Haematology Society pioneered risk- and gender-adapted therapy featuring the vincristine, etoposide, prednisone, and doxorubicin regimen for boys to limit the amount of alkylating agents; girls received vincristine, procarbazine, prednisone, and doxorubicin. The HD-95 trial investigated whether radiation therapy could be omitted in patients achieving a complete remission to chemotherapy. Early results (median follow-up time: three years) indicate a 97% event-free survival rate for favorable-risk patients. There was no difference in outcome between favorable-risk patients treated with chemotherapy alone and those treated with combined-modality therapy [276].

Randomized trials have compared chemotherapy alone with chemotherapy plus low-dose radiation for advanced stage Hodgkin lymphoma. The results have primarily indicated that there is no benefit to the addition of radiation in terms of event-free or overall survival, with one study indicating that the addition of radiation was comparable to the addition of a second chemotherapy regimen [123,277]. Researchers continue to debate the issue of adding low-dose radiation to chemotherapy because of the toxicity associated with each modality [278,279,280]. However, the use of radiation therapy may allow for lower doses of chemotherapy agents. In an effort to cure children with Hodgkin lymphoma with minimal side effects, most treatment approaches entail combined-modality therapy with reduced dose radiation. A number of combined-modality therapy protocols exist, and the decision as to how to incorporate radiation therapy should be made within the context of the protocol followed [272].

Treatment for Relapsed/Refractory Disease

Relapse will occur after first-line treatment in approximately 10% to 20% of children with advanced-stage Hodgkin lymphoma [270]. Relapse occurs most often within four years, but late relapse is not rare [272]. Chemotherapy is the recommended second-line therapy. The choice of specific agents, dose intensity, and number of cycles is determined by the initial therapy, disease characteristics at progression/relapse, and response to second-line therapy [47]. Although chemotherapy with ifosfamide, carboplatin, and etoposide in combination or as single agents is effective, the toxicity profiles of the drugs have led investigators to search for alternative treatments. Vinorelbine, ifosfamide, and high-dose methotrexate have been found to be acceptable, but subsequent relapse after high-dose methotrexate has occurred in as many as 50% of children/adolescents [270]. Nonmyeloablative stem cell transplantation has been used in selected patients with high-risk disease [270]. In addition, two cycles of chemotherapy with dexamethasone, etoposide, cisplatin, high-dose cytarabine, and asparaginase (DECAL) plus maintenance therapy and stem cell transplantation was found to be effective for relapsed Hodgkin lymphoma, with a five-year event-free survival of 26% and five-year overall survival of 31% [281]. The COG has evaluated ABVE-PC (doxorubicin, bleomycin, vincristine, etoposide, prednisone, and cyclophosphamide) and its derivatives across all risk groups [282,283,284]. Benefits of this regimen include lower cumulative doses of anthracyclines, alkylating agents, and bleomycin compared with MOPP and ABVD regimens used in prior decades. Lower cumulative doses are expected to translate into long-term reductions in second malignant neoplasms and fewer cardiovascular, pulmonary, and fertility complications [285,286,287].

COG investigated treatment of relapsed/refractory disease with brentuximab vedotin and gemcitabine in 46 children and young adults [288]. Twenty-four of 42 patients treated at the recommended phase II dose of 1.8 mg/kg experienced a complete response within the first four cycles. Four of 13 patients (31%) with partial response or stable disease had all target lesions with Deauville scores of 3 or less after cycle four. By modern response criteria, these are also complete responses, increasing the complete response rate to 28 of 42 patients. There were no treatment-related deaths. The most common adverse events among all patients treated were neutropenia, rash, transaminitis, and pruritus [288]. Trials are ongoing to determine the efficacy and toxicity of combining brentuximab vedotin with chemotherapy. Additionally, phase I/II trials of immune checkpoint inhibitor therapy (with nivolumab and pembrolizumab) for use in childhood refractory/relapsed disease are ongoing [289,290].

Stem Cell Transplantation

HSCT is usually limited to patients who have relapse or in whom disease is refractory to primary conventional therapy [47,119,252].

Follow-Up and Surveillance

Most pediatric patients with Hodgkin lymphoma have a favorable outcome; however, there is increasing concern about risks of carcinogenesis from diagnostic and therapeutic radiation therapy and from post-treatment surveillance imaging. In one study, cumulative radiation dosage from diagnostic imaging examinations and the frequency of relapse detection by these examinations were recorded. In the first two years after therapy, patients in remission received a median of 11 examinations. In 13 of 99 patients who relapsed, approximately 1% of surveillance imaging examinations (e.g., CT scan, positron emission tomography [PET] scan, chest radiographs) identified relapsed disease. Given this very low rate, the financial burden of the tests themselves, the high cure rate, and risks of second malignancy from ionizing radiation exposure, modification of the surveillance strategy is recommended [291].

Non-Hodgkin Lymphoma

With standard treatment options, approximately 75% to 80% of children/adolescents with non-Hodgkin lymphoma will survive at least five years, although the survival rate varies according to subtype [17,139,292]. Age may also be a factor. In one study, the five-year event-free survival rate was approximately 79% for adolescents compared with 85% for patients younger than 15 years of age [293]. However, another study found that age was not a factor in increased risk of treatment failure [152]. Instead, the patient's level of lactate dehydrogenase, mediastinal disease, and combined bone marrow-/CNS-positive involvement were independent risk factors in children with mature B-cell non-Hodgkin lymphoma [152]. Response to treatment also is an important prognostic indicator. With the exception of anaplastic large-cell lymphoma, pediatric non-Hodgkin lymphoma that is refractory to first-line therapy has a very poor prognosis, regardless of histology [294,295,296].

As with Hodgkin lymphoma, the treatment of non-Hodgkin lymphoma is based on the extent of disease, with the intensity of treatment being increased for more extensive disease [297]. Because non-Hodgkin lymphoma in children is considered to be widely disseminated from the outset, even when apparently localized, combination chemotherapy is recommended for most patients [17]. The event-free survival has been better in some studies in which surgical resection was done. So, there may be some value to this approach if resection can be readily accomplished, especially in cases of large tumor masses [120]. Treatment often consists of a cytoreduction phase to reduce tumor burden, followed by a consolidation phase.

For most studies of treatments done in the United States, patients are categorized into three risk groups: low-risk, which is completely resected stage I disease or stage II abdominal disease; high-risk, which consists of CNS involvement with or without bone marrow involvement; and intermediate-risk, which encompasses disease that is not eligible for the other two groups [120]. Treatment also varies according to the pathologic subtype.

B-cell non-Hodgkin lymphoma can be treated with two or three cycles of combination chemotherapy after surgical resection if there is no measurable tumor burden [17,297,298]. The combination of vincristine, cyclophosphamide, doxorubicin, and prednisone has been highly effective in this setting [221]. A single-agent phase II study of rituximab performed by the BFM group showed activity in Burkitt lymphoma [299]. A pilot study from the COG added rituximab to baseline chemotherapy with FAB/LMB-96 therapy in patients with stage III and stage IV B-cell non-Hodgkin lymphoma [300,301]. For higher risk disease, the addition of two or three courses of chemotherapy, high-dose methotrexate, and high-dose cytarabine was also effective [221,302,303].

Good outcomes for treatment of lymphoblastic lymphoma have been achieved with use of ALL-type regimens consisting of an induction phase of multiagent chemotherapy (eight drugs) followed by a consolidation phase and a maintenance phase of two years, with cranial radiation for CNS prophylaxis [55,304]. This approach was superior to short, intensive pulsed chemotherapy regimens [305,306,307]. Good outcomes also have been reported by the COG with use of a Children's Cancer Group (CCG)-modified BFM ALL regimen (prednisone, dexamethasone, vincristine, daunorubicin, doxorubicin, L-asparaginase, cyclophosphamide, cytarabine, methotrexate, 6-mercaptopurine, 6-thioguanine, and reduced number of intrathecal treatments during maintenance) [17]. The study randomized 254 patients to the regimen with intensified intrathecal methotrexate (Arm A1) or an adapted non-Hodgkin lymphoma/BFM-95 therapy with high dose methotrexate in interim maintenance but no intrathecal methotrexate in maintenance (Arm B1). Each cohort was randomized to either more or less intensification (cyclophosphamide/anthracycline) (Arms A2/B2) and CNS radiation therapy for CNS-positive patients only [308]. There was no difference in five-year event-free survival among the four arms. Five-year survival for CNS-positive patients was 63%. There was no difference in outcome based on CNS prophylaxis or intensification for CNS-negative patients [308]. In a COG study of 56 children and adolescents with stages I and II disease who were treated with the modified CCG BFM regimen for two years, the COG reported a five-year event-free survival of 90% and a five-year overall survival of 96% [309].

For patients with a mediastinal mass, radiation targeting the mass is not necessary except as emergency treatment to relieve airway obstruction or superior vena cava syndrome [17,121]. In such cases, low doses of radiation are used. Cranial radiation may be reserved for patients with CNS involvement at the time of diagnosis [307].

Several treatment strategies have been used for anaplastic large-cell lymphoma, and the outcomes have been comparable [17,122,144]. Treatment has been adapted to risk in some cases, usually according to stage of disease [144]. Common agents are corticosteroids, vincristine, doxorubicin, and methotrexate, but the doses have varied substantially. A course of chemotherapy similar to that used for B-cell non-Hodgkin lymphoma, with treatment stratified by stage of disease, was effective for disseminated anaplastic large-cell lymphoma, as was treatment that included a short cytoreduction phase (two chemotherapy cycles) followed by a short maintenance phase (five to seven months) [143,144]. In an international study of systemic childhood anaplastic large-cell lymphoma, 12 of 463 patients had CNS involvement; three of the 12 had isolated CNS disease. Comparative analysis of CNS-positive and CNS-negative patients showed no difference in anaplastic lymphoma kinase positivity, immunophenotype, presence of B symptoms, or other sites of disease. With multiagent chemotherapy, the event-free and overall survival rates of the CNS-positive group at five years were 50% and 74%, respectively [310]. CNS involvement in anaplastic large-cell lymphoma is rare at diagnosis [17].

CNS Prophylaxis

The use of radiation therapy is limited in children with non-Hodgkin lymphoma. Early studies showed that routine CNS prophylaxis had no benefit for patients with low-risk (stage I or stage II) disease [311]. It has been demonstrated that CNS prophylaxis can be omitted in lymphoblastic lymphoma [307,312]. It can also be eliminated for patients with anaplastic large-cell lymphoma and B-cell non-Hodgkin lymphoma, even those who present with CNS disease [144,302]. These findings are supported by data from the Childhood Cancer Survivor Study, which demonstrated that radiation was a significant risk factor for secondary malignancy and death in long-term survivors [313].

Treatment for Recurrent or Refractory Disease

Relapse occurs in approximately 12% of children/adolescents with non-Hodgkin lymphoma [292]. Intensive chemotherapy with drugs that were not used in first-line treatment have been utilized, and stem cell transplantation support has been effective for patients with chemosensitive disease [120,314,315,316,317]. Two cycles of chemotherapy with DECAL, maintenance therapy, and stem cell transplantation was found to be effective for relapsed non-Hodgkin lymphoma, with a five-year event-free survival of 23% and five-year overall survival of 30% [281].

Although outcomes in pediatric B-non-Hodgkin lymphoma have improved with intensive chemotherapy protocols, the long-term adverse effects of therapy and poor outcomes for patients who relapse continue to be challenges [17,318]. One study evaluated the potential risks and benefits of routine relapse surveillance imaging after the completion of therapy among 44 patients with non-Hodgkin lymphoma who were diagnosed and treated at Texas Children's Cancer Center between 2000 and 2011 [318]. Only three of the 44 patients had relapse (6.8%) and none of the relapses were initially diagnosed by CT or PET scans. The median effective radiation dose per patient was 40.3 mSv (range: 0–276 mSv). This study highlights the low relapse rate following complete response at the end of therapy among children/adolescents with B-non-Hodgkin lymphoma, the low sensitivity of early relapse detection by surveillance imaging, and the costs and potential increased risk of secondary malignancies from cumulative radiation exposure from surveillance imaging [318].

Pain may be related to treatment, diagnostic procedures, and/or progressive disease. The patient's self-report of pain, when possible, is the most reliable indicator [57,354]. Assessment of pain must be appropriate for each child's age, developmental level, and cultural context. Pain assessment tools have been developed for several different age groups among children/adolescents, from birth to 18 years of age, and for nonverbal or cognitively impaired children (Table 16) [57,365,366,367,368,369,370,371,372,373,374]. These tools are designed to either provide a score according to a set of behavioral cues, as in the case of infants and young children, or allow the child to self-report on the intensity and location of pain. Several tools have been modified and validated for use among children/adolescents of different races/ethnicities, including Black, Hispanic, Asian, and Alaska Native populations [375,376,377,378].

It is also helpful to evaluate behaviors to determine if a young child has pain. Indicators of pain in infants include facial expressions such as frowning, a furrowed brow, a quivering chin, crying, sucking, flexing of fingers and toes, and breath-holding. Behavioral indicators are also important for older children, as they may not admit to having pain because of fear of an injection. Indicators in older children include decreased energy level; decreased eating; lack of interest in usual activities; holding or protecting part of the body; seeking comfort or closeness; and whining or groaning [365,366,367,368,369].

Physical examination and functional evaluation are other components of pain assessment. During the examination, the clinician should watch closely for nonverbal cues that suggest pain. These cues are especially important when examining patients who are unable to verbally communicate. To evaluate how pain may be influencing function, the clinician should watch the patient to see how pain limits movements and should ask the patient or family how the pain interferes with normal activities. Determining functional limitations can help enhance patient compliance in reporting pain and adhering to pain-relieving measures, as clinicians can discuss compliance in terms of achieving established functional goals.

As with adults, using the WHO ladder to manage pain is effective for children [379,380,381]. The WHO ladder involves a three-step approach, with the strength of the analgesic agent increasing according to the severity of pain. There are two important underlying principles of the WHO ladder. First, analgesics should be administered on a regular schedule rather than on an as-needed basis. This approach is not only more effective at controlling pain but also avoids unnecessary pain as a prompt for the next dose. The second principle is that treatment should correspond to the intensity of pain as reported by the patient, regardless of whether treatment at a previous step was carried out [379].

The WHO ladder calls for the use of nonopioid and opioid analgesic agents (Table 17). Rotation of opioids has been found to be effective and can help avoid dose-limiting toxicity [178]. Medications for pain should be given by the easiest, least painful route (oral, sublingual, or parenteral for children with a central venous catheter), avoiding intramuscular injections whenever possible [178,354].

There are no randomized controlled trials addressing the management of breakthrough pain in children with cancer. Limited data and considerable experience indicate that breakthrough pain in this patient group is common, underassessed, and undertreated [382]. Clinicians and parents should also be alert to breakthrough pain, which occurred in 57% of children with cancer in one study [383]. The risk of breakthrough pain was higher for younger children (7 to 12 years of age) than older children (13 to 18 years of age). The high rate of breakthrough pain reflects the importance of pain assessment as an ongoing process, with assessment and documentation at regular intervals. The optimum interval varies. In general, patients with more severe disease should be evaluated more frequently. Measures of pain relief should also be documented, especially with regard to efficacy.

An ideal therapeutic agent would be easy to administer and rapid in onset with short duration. The most common and effective strategy seems to be multimodal analgesia that includes an immediate-release opioid (e.g., morphine, fentanyl, hydromorphone) administered intravenously by a patient-controlled analgesia pump, ensuring an onset of analgesic action within minutes. Intranasal fentanyl (or hydromorphone) may be an alternative, but no pediatric data have been published for commercially available fentanyl transmucosal application systems, and these products cannot yet be recommended for use with children with cancer and breakthrough pain [382].

Pain medication should be complemented by nonpharmacologic interventions that are begun early in the course of treatment. These interventions should be age-appropriate. Touch, massage, stroking, and rocking work well for infants, toddlers, and young children; distraction is effective for older children, as are guided imagery, music and art therapy, play therapy, controlled breathing, and relaxation techniques [57].

Fatigue is often described by children as "drowsiness," feeling "sluggish," or having a loss of energy. As many as 70% to 80% of children/adolescents with cancer note one of these feelings, which can be caused by many factors, including medications (chemotherapy agents and opioids), pain, anemia, progressive disease, and psychological issues [348,354,362]. Tools to assess fatigue, for use by patients (7 to 12 years of age), parents, and healthcare staff, have been developed but require evaluation and validation in prospective studies [384,385]. Signs and symptoms of fatigue include asthenia, sleep disturbances, low energy level, and decreased participation in activities. Gradual increases in exercise and school activities and cognitive-behavioral approaches to improve coping skills may be helpful.

One study evaluated the change in children's and adolescents' fatigue scores during cancer treatment and described possible causes of fatigue [386]. Forty children 7 to 12 years of age, 29 adolescents 13 to 15 years of age, and one parent were included in the study. Cancer-related fatigue was measured using the Child Fatigue Scale, the Adolescent Fatigue Scale, and the Parent Fatigue Scale. Children, adolescents, and parents reported a statistically significant increase in fatigue scores during their treatment, with medical procedures and the hospital environment reported as major causative factors of the fatigue experienced [386]. Results of another study indicate that cancer-related fatigue results in a low health-related quality of life for pediatric patients [387].

Nausea and vomiting are the most common gastrointestinal symptoms in the pediatric cancer setting; other symptoms include constipation, diarrhea, and anorexia/cachexia. Nausea and vomiting can be a challenge to assess, especially in young children who are unable to articulate the feeling of nausea. Inactivity, signs of weakness, and/or lack of appetite may indicate nausea. The choice of antiemetic is based on the likely cause of nausea and vomiting [178,354]. The best agents for nausea and vomiting caused by chemotherapy or radiation therapy are 5-HT3 receptor antagonists, such as ondansetron and granisetron. Other choices are prochlorperazine, scopolamine, dronabinol, and metoclopramide [119,178,354,388]. Anticipatory nausea may be alleviated by benzodiazepines and the use of relaxation techniques [329,354]. Use of a hydration fluid may be necessary to prevent dehydration in a child who is vomiting excessively [354]. Taking medications after meals, when possible; eating small, frequent meals; and avoiding strong smells may also help ease nausea [178].

Constipation is also common among children/adolescents with cancer and is usually related to medications, particularly opioids and vincristine [354]. This symptom is often under-recognized or undertreated by clinicians [178]. The optimum treatment for constipation is prevention through a variety of measures, including movement/exercise, adequate fluid intake, and increased fiber intake. Regular use of laxatives is recommended during treatment with opioids [57].

Mucositis occurs in as many as 50% of children/adolescents receiving chemotherapy. It is often associated with radiation therapy to the head and neck area [329,343]. Mucositis can affect the mouth, esophagus, stomach, duodenum, and colon, and the severity depends on the type of chemotherapy used. High-dose methotrexate and high-dose cytarabine are considered the most damaging to mucosa [343]. Small studies have shown that chlorhexidine is beneficial for alleviating mucositis in children, but the results must be confirmed by larger trials [389]. One literature review focused on randomized clinical trials to evaluate the effectiveness of chlorhexidine mouthwash in alleviating mucositis in children receiving chemotherapy [390]. Control groups consisted of placebo, no intervention, or another intervention group. Mucositis was scored using either the WHO scale or the modified Oral Assessment Guide. Five studies fulfilled the inclusion criteria, in which chlorhexidine was evaluated. Four of the five showed a significant preventive effect on the development or severity or mucositis when chlorhexidine mouthwash was used. The remaining study showed no benefit compared with the placebo group [390]. Other helpful measures include the use of ice chips during administration of chemotherapy, excellent oral hygiene, analgesic agents, and avoidance of irritating foods or fluids, such as citrus fruits and juices [329,343].

Pruritus usually occurs as a side effect of medications, especially morphine and other opiates, and may also be associated with progressive lymphoma [57]. Physical manifestations of pruritus include excoriation, erythema, and lichenification; rubbing of the eyes, nose, or skin may also indicate pruritus [256]. Discontinuing use of the causative medication, if possible, is the preferred treatment approach. Moisturizing creams may be used for dry skin, and systemic corticosteroids or antihistamines may be helpful for more severe cases [256].

Bereavement and grieving support are other essential components in end-of-life care [178,352,419]. Bereavement care may best address grief by providing family members with outlets to communicate feelings and memories about a dying or deceased child [422,423]. It is important to provide support for siblings, who report often feeling left out and abandoned [352,419]. Despite a growing body of literature substantiating the needs of siblings, they continue to be unmet [424,425,426,427]. Siblings also may experience post-traumatic stress symptoms, poor quality of life, and a sense of aloneness [428]. Healthcare providers can address the needs of siblings by [428,429,430,431]:

Grief can be prolonged and exacerbated by several factors, including the loss of the child's quality of life, the family's dissatisfaction with the provider-family rapport and communication, cultural insensitivity, and lack of follow-up [178]. This finding underscores the need for effective end-of-life care and open, honest, empathic communication from the palliative care team. In addition, caregivers should respect the family's cultural context when designing and discussing treatment plans and goals. Also, it is imperative that family members be allowed to hold their child during and after death and to carry out family, religious, and cultural rituals.

In most cases, the grieving process is considered to be in its end stages when bereaved individuals are comfortable with a restructured lifestyle [432]. However, parental grief related to the loss of a child is generally experienced more intensely and differently than other forms of grief because a child's death is atypical in developed countries [433]. Parents can maintain healthy bonds with their deceased children by sharing memories about their children, writing biographies, performing annual religious rituals, and establishing monetary memorials [434,435]. The maintenance of the connection through conversation appears to aid in healing for grieving parents [435].

Grieving is relieved by the continued involvement and support of the palliative care team and the primary care provider after a child's death [436]. Attendance at a memorial service and handwritten notes of sympathy from the clinician and medical team are valued highly by grieving families [429,437]. The clinician should emphasize the personal strengths of the family that will help them cope with the loss and should offer help with specific issues. The note should also invite the patient to contact the physician or other members of the palliative care team with questions. Provision of bereavement services varies. Programs usually involve contacting the family at regular intervals to provide resources on grieving, coping strategies, professional services, and support groups. Frequent contact with family after death can ensure that families are adjusting to the loss [352]. Referrals for psychosocial and spiritual interventions should be made as early as possible to optimize their efficacy. Bereavement services should extend for at least one year after the patient's death, but a longer period may be necessary.

Coping has been defined as the "constantly changing cognitive and behavioral efforts to manage the specific external and/or internal demands that are appraised as taxing or exceeding the resources of the person" [441]. The two main types of coping are problem-focused and emotion-focused [441]. The goal of problem-focused coping is to change aspects of an event to relieve distress. If the individual cannot manage a stressful event, emotion-focused coping is used to regulate feelings or alter his or her interpretation of the situation. Emotion-focused coping mechanisms intended to alter interpretations are categorized into four control strategies: predictive control, vicarious control, illusory control, and interpretative control [360]. These control strategies are used commonly by children and their parents when coping with cancer [361,362]. Using combinations of problem-solving and emotion-regulating mechanisms to meet each cancer-related challenge decreases short-term distress and promotes healthy social and cognitive development and adaptation [442].

The majority of studies have shown that the psychological profile of childhood leukemia or lymphoma survivors is similar to population norms, suggesting that most children with leukemia or lymphoma will cope effectively with their cancer experience [443,444]. However, subgroups of these patients are at special risk for poor psychological outcomes [445,446,447].

As children learn and grow, they construct the world around them differently. Thus, the developmental age at the time of a cancer diagnosis defines the type and complexity of coping mechanisms available to the patient throughout the course of the disease [448,449]. The level of understanding and stressors, coping mechanisms, potential long-term effects, and actions a medical team can take to promote well-being vary according to the developmental age [450,451].

Infants (Birth to 12 Months of Age)

Level of understanding and stressors: The infant's understanding of treatment is limited to pain and comfort. Infants undergoing cancer treatment can have a difficult time developing trust and bonding with parents because medical procedures often separate infants from their parents, disrupt household routines, and cause pain. An infant frequently separated from parents may experience feelings of abandonment [452].

Coping mechanisms: Infants are limited to certain aspects of emotion-focused coping because crying is an emotional release. With few available coping mechanisms, the infant's emotional development and future coping abilities can be affected adversely by cancer.

Potential long-term effects: A child diagnosed with cancer during infancy may experience cognitive delay. Difficulty bonding with parents may affect subsequent developmental stages.

Actions to promote well-being: To offset cancer-related developmental disruptions, the medical team should allow parents to stay in the infant's room and attend to his or her needs, such as feeding, diaper changing, comforting, and bathing. In addition, minimizing the number of staff caring for the child may promote trust and a sense of routine.

Toddler (18 Months to 3 Years of Age)

Level of understanding and stressors: A toddler diagnosed with cancer has different developmental challenges than an infant. However, a toddler still has limited understanding of cancer-related cause and effect and has limited access to coping mechanisms. Because a toddler's sense of time is restricted to the present, the child will likely not grasp that staying still reduces the time and pain associated with treatment. The toddler's level of understanding may exceed his or her ability to communicate, which is another barrier to coping. During this stage of development, a toddler explores his or her sense of self, property, and security. Cancer therapy, however, challenges a toddler's autonomy and feeling of safety. For instance, a toddler will likely say "no" to treatment, only to be vetoed. In addition, the toddler may perceive that his or her parents cannot protect him or her from pain associated with treatment. Although a toddler's understanding of death is minimal, he or she likely has some concept of loss and separation and may fear abandonment when parents are out of sight. The health-related quality of life for toddlers with cancer is significantly worse than that of healthy age-matched controls. Cancer most strongly affects a toddler's sleep, appetite, behavior, and liveliness [453].

Coping mechanisms: A toddler has access to more emotion-focused coping strategies to regulate stress than an infant and has developed some problem-focused strategies. Communication skills, although limited, can be used to express emotions and seek parental support and solace. To gain more control over his or her environment, a child may hold a parent's hand during a painful procedure to know that the parent is near. Pain endured during a procedure may be associated with the staff member providing the respective treatment.

Potential long-term effects: During this stage, a cancer diagnosis may postpone language and motor development as well as bladder and bowel control. Toddlers' development in these areas may regress.

Actions to promote well-being: Parental handling of a toddler's cancer-independent needs should be made routine to minimize the child's feelings of abandonment or insecurity. Motor skill development accelerates at this age; therefore, safe levels of physical activity should be promoted. Also, giving a toddler limited choices, even in the smallest matters, supports their sense of autonomy. For instance, a provider could ask a toddler, "Would you like a bandage with cartoon characters or a regular one?"

Early School-Age (4 to 7 Years of Age)

Level of understanding and stressors: Although the degree of development of a child at 4 years of age may be much different from that of a child at 7 years of age, children of these ages have surpassed the cognitive and emotional development of toddlers. Concepts of empathy and guilt are developing. Injury and death may be believed to be magically reversible, as in many children's cartoons. Patients who are 6 to 7 years of age at the time of a cancer diagnosis have been shown to have poor health-related quality of life with respect to autonomy, motor skills, and cognitive functioning two months after treatment [453].

Coping mechanisms: Early school-age children can cope using emotion-focused methods and some problem-focused strategies. A child in this age category can communicate feelings and needs more clearly than a toddler and has a better understanding of cause and effect. Both of these factors elevate the child's ability to cope actively with stressors and express negative emotions. Children understand the concept of hopefulness during this stage, which can promote a positive outlook and adaptive coping [454]. Regardless of the cancer stressors experienced, patients with a high sense of humor have better psychosocial adjustment and fewer infections [455].

Potential long-term effects: Experiencing leukemia can interfere with the proper development of an early school-age child's social and academic skills [456,457,458,459]. Without honest, age-appropriate discussions about the child's prognostic information and death, the development of thinking based on logic rather than fantasy may be delayed [460]. Children in whom cancer is diagnosed after 5 years of age are more likely to experience post-traumatic stress symptoms because they are able to understand the seriousness of their illness. However, they are also more likely to experience post-traumatic growth, characterized by a positive shift in perspective and priorities, because they can appreciate their inner strength and others' support [461].

Actions to promote well-being: A child may believe that his or her cancer diagnosis and treatments are punishments for bad behavior. Parents should be advised that rewarding good behavior during and after procedures with praise and small rewards (e.g., stickers) can promote a child's healthy development and ability to cope with cancer. Negative behaviors should not receive extra attention. To avoid developmental delay of reality-based thinking, parents should also be advised to answer a child's questions about the cancer honestly and in an age-appropriate manner [460]. Role-play discussions should be performed to assist in understanding how to best tell an individual child distressing news [185]. Discussions should be tempered by the expression of hope, which is a very important coping strategy in any pediatric cancer situation. School attendance should be encouraged because social and academic skills can be developed and peers can serve as an important support system for children with cancer [457]. Animal therapy programs may reduce a young child's emotional distress and promote socialization [462].

School-Age (8 to 12 Years of Age)

Level of understanding and stressors: Compared with self-reports in other developmental age groups, self-reports of children with cancer who are 8 to 12 years of age and secondary assessments were more likely to emphasize emotional distress over physical distress. Emotional stressors include confinement, feelings of alienation, and worries before medical procedures. The pain associated with diagnostic procedures and treatments is the most frequently mentioned physical aspect of cancer-related distress [463]. By 9 years of age, children understand that death is irreversible; however, their concept of the future is still very limited [464]. Some children may try to "protect" their parents from fearing death by avoiding conversations about the subject. However, avoidance of these discussions causes a child to suffer in isolation [190,460].

Coping mechanisms: Communication and logical reasoning skills available to children in this age range promote the development of problem-focused coping. Children may be more interested in learning about their cancer and its treatment because they want to master social and cognitive skills; intellectualizing the problem is one coping strategy children of this age may use. Emotion-focused coping skills are also more advanced because a school-age child is more likely to communicate emotions to seek others' support [463].

Potential long-term effects: Intensive treatments may interrupt the development of social and academic skills. Although the child's communication skills are more sophisticated, sharing concepts that are "too adult" for the child can affect socialization with peers and personality development.

Actions to promote well-being: A cancer diagnosis poses a challenge to a child's desire for equality and fairness. Its effects on schoolwork and physical activity may leave the child feeling inferior to peers, angry, and frustrated [457]. Such emotions may manifest as withdrawal or depression; the child may hide these feelings from parents. Thus, listening carefully and discussing feelings is imperative to understanding the child's problems. Art, writing, music, and dance/movement therapy can provide the child with expressive outlets [465,466,467]. Although children may desire to know all about their illness, parents and the treatment team must provide age-appropriate education. Too much exposure to decision making and adult responsibilities may disrupt a child's socialization with peers. As for any school-age child, school attendance is beneficial for social and academic development, and children should therefore be given the tools necessary to transition back into a school routine [457].

Adolescent (13 to 18 Years of Age)

Level of understanding and stressors: During this stage, adolescents are attempting to individualize themselves. However, identity formation in adolescents with cancer may revolve around their illness. Although adolescents understand death, they may feel it only happens to others. The most frequently reported emotional stressors are worries before medical procedures and feelings of alienation [363,463,468]. Compared with age-matched controls, children 12 to 15 years of age with cancer have a significantly lower health-related quality of life with respect to motor functioning [453].

Coping mechanisms: Coping strategies commonly used by adolescents during treatment include positive thinking, avoiding thoughts of treatment, keeping busy, focusing on positive outcomes, and staying relaxed [468,469,470]. One study found that better problem-solving skills were associated with lower adolescent distress [471]. A systematic review analyzed how childhood cancer patients and survivors communicate about their cancer with family and peers [472]. Participants were adolescents and young adults 16 to 34 years of age. Communication occurred on a spectrum with variations in who the youth shared information with, how often they shared information, and the amount and type of information they shared. The reviewers found that communicating about cancer was a beneficial coping strategy and that it prompted social support and appeared central to significant relationships. Barriers to communication (e.g., fear of stigma, poor peer reactions) hindered the participants' willingness to disclose [472]. Post-treatment coping strategies include negotiating to regain/gain autonomy, cognitive reliving of the cancer experience with peers or the medical team, and forgetting the negative aspects of treatment [469]. Potential long-term effects: One large, retrospective study has suggested that adolescents with cancer achieve autonomy, psychosexual development, and social development at an older age than peers [473]. Diagnosis at adolescence has been linked to an increased risk for the development of post-traumatic stress disorder [446]. However, one study found that personality factors, such as adaptive style, were a stronger determinant of post-traumatic stress than health history [474]. Another study of 435 youth (253 with cancer, 182 healthy controls) found that the relationship between post-traumatic stress symptoms and post-traumatic growth depended on contextual factors and that the majority of the youth (83%) were resilient and able to find benefit from stressful life events [475].

Actions to promote well-being: The physical and emotional changes of this final stage of childhood are conjoined with a desire for private space. However, honoring an adolescent's desire for privacy can be difficult during hospitalizations. A sense of privacy and respect, as well as cooperation, can be maintained by knocking before entering an adolescent's room, asking permission when introducing new people, and being mindful of the adolescent's physical modesty during examinations. To promote treatment compliance in older adolescents, equal medical information should be provided to patients and their parents. Adolescents should have the option to lead treatment discussions in order to enhance feelings of respect, control, and independence. Getting to know others with the same disease, treatment, or situation—especially those who are cured—may help adolescents cope, and emerging technology, cancer-related video games, and social networking are helping adolescents with cancer to better understand their disease, comply with treatment, and connect with one another [9,468,476]. Parents should be dissuaded from controlling treatment discussions and decision making without the adolescent's approval, which can lead to acting-out (externalizing) or internalizing behaviors in the patient.

Gender Differences in Coping Mechanisms

Some studies have shown differences in the coping styles of male and female children/adolescents with cancer. One study found that older female children (12 to 18 years of age) use a wider repertoire of emotion-focused and problem-focused coping strategies than their male counterparts [477]. Among male participants in the study, use of emotion-focused coping was linked to a disposition to experience anger and anxiety but not depression. Emotion-focused coping in the female participants was not associated with these variables. Depressive symptoms were less likely to develop in female participants who used problem-focused strategies.

In another study, male children 11 to 18 years of age were found to use disengagement coping strategies, such as problem avoidance, wishful thinking, social withdrawal, and self-criticism, more often than female children did [478,479]. Engagement coping strategies include problem-solving, cognitive-restructuring, social support, and expression of emotions. Male children were also found to use humor coping significantly more often than girls in a study of children 7 to 14 years of age who had ALL [455,480]. Healthy psychosocial adjustment was strongly correlated to humor coping [455].

Mothers are the most commonly reported source of social support by children/adolescents with cancer. As a child with cancer develops into an adult, however, his or her social support needs extend beyond the family. Socialization with peers contributes to coping abilities and psychological adjustment of children/adolescents with cancer [481,482,483]. For adolescents 12 to 14 years of age, family-independent social networks are significantly smaller than those of adolescents 15 to 18 years of age, which may contribute to less effective coping. The emotional support of peers is particularly important for continued participation in school and social activities [483]. Hospitalizations, poor physical health/appearance, academic difficulties, and poor parental coping skills can also limit the patient's development of relationships with peers [457,484].

Regular attendance at school is difficult during cancer treatment, but clinicians should emphasize to patients and their parents that going to school is an important part of normal life. School can help to reduce the social isolation associated with cancer and enhance the child's self-esteem [320]. School reintegration programs have been found to improve children's psychosocial well-being (anxiety and depression), social well-being (social competence and social support), behavioral problems, and physical competence, but more research is needed to investigate the effects of these and other related interventions [192,485].

In preparing for the child/adolescent to re-enter school, the primary care provider should guide parents in developing questions and issues to discuss with teachers and other personnel at the school. Topics that should be addressed with school personnel include [486]:

Cancer and its treatment are often associated with learning problems, and parents should be aware for the potential for difficulties with handwriting or spelling, reading comprehension, remembering math facts, copying information after seeing it, and completing tasks on time [486].

A meta-analysis of the effects of psychosocial interventions on psychological outcomes in children with cancer showed that patient-oriented therapies do not effectively reduce child distress or improve child adjustment [487]. The authors of the study suggested several reasons for this finding: children with cancer have psychological outcomes similar to those of healthy children; the average length of follow-up of the studies analyzed was only four months; and most studies used a waitlist/standard care control group instead of a control group with a generic psychosocial intervention providing nonspecific support or attention. Waitlist control groups do not control for nonspecific treatment variables such as social support from other group members or facilitators. One systematic review identified a broad range of psychosocial interventions that could benefit cancer survivors and their families, but concluded that more high-quality studies are needed to optimize the healthcare services that can best support children and their families [488].

Parental factors linked to the psychological outcomes and quality of life for a child with cancer include education, communication styles, and coping strategies [457,478,485,489,490,491,492,493].

From the diagnosis through the course of a child's cancer, cancer-free siblings of all ages are affected emotionally and often physically. Siblings may suffer from unattended emotional responses, including anger, anxiety, jealously, resentment, depression, and loneliness; difficulties related to the absence of their parents; and reduced school achievement [494]. Immature thinking in young children can lead to thoughts that they caused their sibling's illness. Adverse physical problems experienced by cancer-free siblings are headaches, sleep disruption, and poor eating habits [495]. Most siblings who experience these emotional and physical responses do not suffer long-term problematic outcomes; however, a subset of siblings does experience difficulties [496].

Risk Factors for Poor Psychological Outcomes for Cancer-Free Siblings

A meta-analysis of 50 studies involving siblings of children with chronic illness showed that internalizing behaviors, especially anxiety and depression, placed healthy siblings at a higher risk of poor psychological adjustment [497]. The risk of unhealthy psychological adjustment is also amplified if the ill sibling's cancer alters day-to-day family function. Warmth or closeness between cancer-free siblings (8 to 15 years of age) and their siblings with cancer increases the risk of poor psychological adjustment; these intimate relationships are associated with decreases in social competence [498]. In addition, the incidence of poor psychological outcomes in response to guilt and heightened fear of death increases if the sibling with cancer dies [499]. The psychological adjustment of cancer-free siblings, however, improves with age as a result of the development of protective factors such as increased maturity and capacity to empathize [498,500].

A post-traumatic stress model may best fit the potential psychological outcomes of cancer-free siblings up to five years after treatment. As a component of the Surviving Cancer Competently Intervention Program randomized trial, 78 sibling participants completed a Post-traumatic Stress Disorder Reaction Index [501,502]. Post-traumatic stress symptoms were experienced by almost half of cancer-free siblings. Moderate-to-severe post-traumatic stress disorder was reported by one-third of cancer-free siblings. Cancer-free siblings who were 7 years of age or older at the time of their sibling's cancer diagnosis reported more severe post-traumatic stress symptoms than those who were 6 years of age or younger at the time of diagnosis. In the same study, cancer-free female siblings indicated more post-traumatic stress symptoms than cancer-free male siblings.

One study explored bereaved siblings' advice to healthcare professionals working with children with cancer and their families [503]. Of the 174 siblings that participated, 108 answered an open-ended question about what advice they would give. The most common advice (given by 56% of siblings) related to their own support. Another one-third wanted professionals to provide them with better medical information. Some siblings wanted to be more involved in their brother's/sister's care and asked that healthcare professionals provide their parents with guidance on how to achieve this. Other common threads were related to psychosocial aspects, for example, siblings' wish for professionals to mediate hope without sacrificing realism and understanding the importance of asking the ill child about the kind of care they wanted [503].

Factors that promote healthy psychological outcomes in cancer-free siblings are related to support and functioning of the entire family. Specifically, these factors include [504,505,506]:

Cancer-Free Siblings' Feelings During Treatment

Cancer-free siblings tend to identify themselves through the experiences of other family members during the course of their ill siblings' cancer [507]. When encouraged to discuss their personal experiences and feelings, cancer-free siblings focused primarily on their ill siblings' or parents' experiences, diminishing their sense of self. These patterns were more evident during the ill siblings' hospitalizations and as the physical appearance of their ill siblings worsened. Although cancer-free siblings experience stress similar to their brothers or sisters with cancer, most participants reinforced that their suffering was not justified because they were not experiencing the illness. Healthy siblings worried about the physical and emotional needs and abilities of their ill brothers or sisters following a cancer diagnosis. They said that they often "felt good" during the times they helped the ill sibling feel more comfortable. Yet, most of the cancer-free siblings also indicated that they experienced guilt because they had minimal involvement in taking care of their ill sibling. Although feelings of guilt in cancer-free siblings have not been correlated to negative psychological outcomes, these feelings should be addressed by parents and clinicians [507].

Numerous studies have shown that cancer-free siblings of children with cancer have the most unmet emotional needs relative to all other members of the family [508]. For children, the family is the primary social support system, and cancer-free siblings receive the least attention when childhood cancer disrupts everyday household life. Cancer-free siblings' emotional responses (e.g., fear, anger, guilt, jealousy, sadness) and health-related responses (e.g., sleeping difficulties, eating problems) are often overlooked or underestimated by parents [495]. These responses are compounded by cancer-free siblings' sensitivities to the parents' cancer-related emotional and practical burdens.

Parental and Self-Reports of Cancer-Free Siblings' Health

Parental and self-reports regarding cancer-free siblings' emotional and physical states also differ. Parents generally report better statuses for their cancer-free children than the children report themselves [509]. However, parents of children with cancer also report that their cancer-free children are "less healthy" than age-matched and sex-matched controls. Despite these reports, parents utilized health care for their cancer-free children less than the parents of matched controls [495]. The relative health of a cancer-free sibling to that of a child with cancer may cause parents to minimize the significance of the cancer-free sibling's health issues. Lack of time, energy, and financial resources may also reduce the parent's efforts to seek health care for a cancer-free child. Furthermore, most cancer-free children attempt to protect their parents' emotions by hiding their own negative emotions and health issues. The findings of a qualitative study suggested that cancer-free siblings valued keeping the family together and the survival of the sibling with cancer similarly [507].

Compared with matched controls, cancer-free siblings of children with cancer report significantly more trouble sleeping (72% compared with 53%) and problems eating (42% compared with 26%) [495]. These complaints are reported more frequently by cancer-free siblings who also report internalizing behaviors than those who engaged in externalizing coping strategies. Cancer-free siblings who present emotional or behavioral problems before the cancer diagnosis are more prone to somatic symptoms and poor psychological adjustment. Therefore, healthcare practitioners should ask parents if their cancer-free children acted-out or had emotional problems prior to the development of cancer in the ill sibling to help assess the cancer-free child's level of risk for psychosocial issues. Also, clinicians should ask if the cancer-free sibling has any physical complaints in order to evaluate whether he or she needs additional interventions to avoid long-term psychological repercussions. Parents who understand that their cancer-free child is at risk for psychosocial problems due to a sibling's cancer are more likely to have the child participate in a therapeutic program [509].

Although parental emotional distress and psychosomatic complaints are high shortly following a child's cancer diagnosis, prospective studies have shown that most survivors' parents adjust within 1 to 10 years [510,511,512]. Still, approximately 25% to 33% of parents face long-term psychological issues [365,510,512,513]. Healthcare professionals' attention to the specific needs of the family can enhance the family's resilience [514,515,516,517]. Post-traumatic stress disorder is the most common psychological disturbance for parents of childhood cancer survivors [513,518].

In a prospective and longitudinal study to assess psychological functioning, parents' somatic symptoms and feelings of anxiety gradually decreased to control group levels by five years after diagnosis [519]. However, 27% of parents still had clinically elevated scores for psychological distress at that time. When the data were analyzed according to the treatment outcome for the child, psychological distress was manifested in 23% and 29% of parents of survivors and deceased children, respectively. Fifty percent of parents of children with relapsed disease continued to experience psychological distress after five years. Higher levels of psychosomatic complaints at the time of the cancer diagnosis and the child's relapse were indicators of higher long-term psychological risk. The same study found that mothers had higher levels of anxious feelings than fathers; however, all other measures were similar between mothers and fathers. Support-seeking coping strategies have been found to significantly reduce anxiety levels in mothers [520].

Therapies that holistically address the family's or the parents' emotional distress and coping skills are most effective for curtailing the intensity of unhealthy psychological outcomes caused by childhood cancer [487,521,522]. Of interventions that have been studied in randomized trials, two have been shown to offer significant benefit [487]. Mothers of children recently diagnosed with cancer were better able to cope with their cancer-related stressors after receiving problem-solving skills training in conjunction with standard psychosocial care, compared with mothers who received standard psychosocial care alone [521]. The effectiveness of the training was better for Spanish-speaking mothers than English-speaking mothers, and single mothers derived the greatest benefit from the training.

The second intervention with proven efficacy is the Surviving Cancer Competently Intervention Program, a one-day, four-session, family-centered treatment intervention designed to teach coping skills to mothers, fathers, and adolescent survivors [522]. Compared with the control group, fathers derived the most benefit from the program, followed by the adolescent survivors. The coping strategies taught to fathers and survivors allowed them to reframe their reactions to the cancer and to speak more openly with their families about the cancer experience. Mothers did not show improvements with the intervention.

Interventions for siblings have also been effective. A therapeutic peer support camp was shown to improve mental health outcomes for siblings (8 to 13 years of age) of children with cancer [523]. The camp program was designed to decrease stress levels, improve social competence, and enhance knowledge of cancer and its treatment. According to standardized self-report measures, participants in the program demonstrated significantly decreased anxiety, improved social competence, greater social acceptance, and significantly reduced fear of disease [523]. Another study found that structured teaching and psychosocial sessions at a five-day residential summer camp led to significant improvements in knowledge about the sibling's cancer, behavior, social support, self-esteem, attitude, and mood [524]. The improvements represented increases of 5% to 25% over the baseline values and were sustained over a period of one year. One study surveyed children's pain and distress levels immediately before and after a one-hour Kids Kicking Cancer in-person class [525]. Eligible participants were enrolled in standard classes, diagnosed with a chronic health condition (e.g., cancer), or the sibling of a child diagnosed with a chronic health condition, and between 5 to 17 years of age. Pain and distress were reported using Likert-style scales. Friedman test was used to test for overall changes in pain and distress, and within subgroups. A total of 59 youth (19 cancer patients, 17 non-cancer patients, 23 siblings) completed the study. Overall, pain and emotional distress were significantly reduced following a one-hour class, with 50% and 89% of youth reporting a reduction in pain and distress, respectively. The reduction in pain was most pronounced with cancer and non-cancer patients, whereas, the reduction in distress was most pronounced among healthy siblings. Reductions in pain and distress did not significantly differ among subgroups [525].

Resources for Patients and Their Families

Several resources are available to help patients and their families become more knowledgeable about cancer and treatment options, gain support from other families, and obtain help with practical concerns. Educational materials developed by several advocacy organizations are available online, and a list of potentially useful resources is available at the end of this course.

The occurrence of subsequent neoplasms as a late effect among childhood cancer survivors has been well-documented. Multiple studies have demonstrated that these cancers are a major source of morbidity and mortality for 30 years or longer after treatment [3,273,526,544,545,546,547,548]. The 30-year cumulative incidence is 20.5% for a subsequent neoplasm and 7.9% for a second malignant neoplasm (excluding nonmelanoma skin cancer) [7]. The highest cumulative incidence is associated with Hodgkin lymphoma (18.4%); the incidence is similar for non-Hodgkin lymphoma and leukemia (5.8% and 5.6%, respectively) [7].

The greatest risk factor for a subsequent neoplasm is radiation therapy (relative risk: 2.7), followed by female sex and childhood Hodgkin disease (relative risk: 1.5 for both), and older age at the time of childhood cancer diagnosis (relative risk: 1.3) [7]. It is important to note, however, that changes to treatment protocols over time (e.g., the move to limited use of radiation therapy) may affect risk among children/adolescents treated within the past decade. Overall, the most frequently occurring subsequent solid tumors among childhood cancer survivors are breast, thyroid, and CNS cancers and soft-tissue sarcoma [7,543].

The authors of one study sought to find the incidence and excess risk for breast cancer in women after chest radiation; differences between the clinical characteristics of breast cancer and the outcomes after therapy among these women and in the general population; and potential benefits and harms associated with breast cancer surveillance among women exposed to chest radiation. The cumulative incidence of breast cancer by 40 to 45 years of age ranged from 13% to 20%. The risk for breast cancer increased linearly with radiation dose. The evidence suggests that the characteristics of breast cancer in these women and the outcomes after diagnosis are similar to those of women in the general population [542]. Authors of another study evaluated cumulative breast cancer risk in 1,230 female childhood cancer survivors treated with chest radiation. Survivors who were treated with lower delivered doses of radiation (2–20 Gy) to a large volume (i.e., whole-lung field) had a high risk (43.6%) of breast cancer. The cumulative incidence of breast cancer by 50 years of age was 30%, with a 35% incidence among Hodgkin lymphoma survivors. Breast cancer-specific mortality at 5 and 10 years was 12% and 19%, respectively [540].

The rate of cancer screening among childhood cancer survivors has been lower than optimal, and clinicians should emphasize the importance of screening, given the high risks [549]. Second cancers among childhood cancer survivors call for vigilant screening and surveillance according to individual risk [534,550]. A prospective one-arm study was conducted among a random sample of 72 Hodgkin lymphoma survivors, 27 to 55 years of age, who were at increased risk for breast cancer and/or cardiomyopathy and had not had a screening mammogram or echocardiogram within the prior two years. A one-page survivorship care plan with recommendations for surveillance was mailed to participants. In addition, survivors' primary physicians were contacted and provided patient-specific information. A web-based virtual information center was made available for both survivors and physicians. Outcomes were assessed by telephone six months after the intervention. At six months, 41% of survivors reported having completed the recommended mammogram; 20% reported having an echocardiogram (women 30%, men 10%) [543].

The greatest risk of cardiac abnormality is associated with a high cumulative dose of anthracycline, which is used in treatment regimens for approximately 60% of children and adolescents with cancer [217,218,526,541,551]. In a study of more than 1,300 childhood cancer survivors, the relative risk of cardiac abnormality was reported to be 3.53 in association with use of an anthracycline and 7.41 in association with the combination of an anthracycline and an alkylating agent [4]. The risk of anthracycline-induced cardiac abnormality is increased further by several factors, including female sex, younger age (younger than 5 years) at the time of treatment, and Black race [534]. Examples of nonanthracycline agents used in pediatric populations that have been associated with cardiotoxicity include cyclophosphamide, cytarabine, cisplatin, and ifosfamide [552].

A wide range of cardiac abnormalities have been noted in survivors [3,4,216,319,526]. It is well-established that the development of congestive heart failure can occur at any anthracycline dose, but the risk increases with increased cumulative doses, especially doses ≥300 mg/m² [553,554,555,556]. A systematic review indicated a rate of subclinical cardiotoxicity of approximately 16% to 28% associated with an accumulated dose of anthracycline of more than 200 mg/m² [4]. Among survivors who had been treated with an anthracycline, cardiomyopathy has been identified in as many as 57% and clinical heart failure (usually asymptomatic) in as many as 39% [216,526]. In a study of more than 10,000 survivors, the relative risk of congestive heart failure was 15.1 compared with sibling controls [3]. The more time that has passed since a patient has received treatment with an anthracycline, the higher the patient's risk for developing changes in cardiac function [217,557,558].

Radiation therapy to the thorax has also been associated with cardiac damage, including decreased left ventricular mass, end-diastolic wall thickness, valve defect, and early, significant decline in left ventricular shortening fraction, with elevated body mass index and Hispanic ethnicity identified as independent risk factors [216,526,541]. The use of lower doses of radiation for children with lymphoma has decreased the occurrence of radiation-related cardiac toxicity [216].

The COG guidelines suggest yearly history and physical examination for survivors at risk for cardiac abnormality related to either chemotherapy or radiation [534]. In addition, an echocardiography and electrocardiography should be done at baseline (the beginning of long-term follow-up) and periodically, depending on risk. Survivors should be educated about the risks of strenuous sports or exercise.

The most common abnormalities in growth and development among childhood cancer survivors have been impaired growth (short stature) and obesity.

Impaired Growth

The use of chemotherapy and radiation therapy in children has been associated with impaired growth and short stature as a result of effects on growth centers in the spine and long bones and endocrine organs [273,559,560,561]. Cranial radiation at a dose of more than 18 Gy has been linked to the greatest loss in height. This loss can be caused by several factors, including growth hormone deficiency, precocious or delayed puberty, or hypothyroidism [273,319,534]. Other factors, such as age younger than 4 years, female sex, obesity, concomitant use of corticosteroids, and radiation dose, contribute to risk [273,534,559].

Cranial radiation alone is not the only causative factor, as final height has been compromised in children with ALL regardless of whether radiation was given in addition to chemotherapy [559,560]. Growth is impaired primarily during remission induction therapy, and so-called "catch-up growth" will occur in approximately 70% of children after induction therapy has been completed [526,559,560,561]. However, maintenance therapy with methotrexate and 6-MP may prevent catch-up growth [562]. Catch-up growth does not occur with radiation-induced impairment [560].

Radiation therapy to the head and neck or cerebral cortex can cause hypothyroidism, leading to short stature and poor growth [273,563]. This risk has been somewhat decreased by the use of lower cumulative doses of radiation therapy in treatment protocols for childhood lymphoma [216]. Bone marrow transplantation is also associated with a high risk for severe growth impairment, especially when the conditioning regimen involves total body radiation [273].

Survivors should have regular growth assessment consisting of measurement of height on standardized growth charts [534]. Growth hormone therapy may be beneficial if growth hormone deficiency is found to be the cause of impaired growth [319,526,534]. Although the use of growth hormone therapy was once thought to confer an increased risk of second neoplasms, evidence of this risk has not been demonstrated in large studies [273,319].

Obesity

Obesity is another well-recognized long-term complication in survivors of ALL, occurring in as many as half of long-term survivors [273,319,526]. Weight gain is increased the most during the first year after the end of treatment, and gains continue with time. Several factors have been suggested as the cause of obesity, including treatment with corticosteroids (especially dexamethasone) and cranial radiation [273,526,534].

A meta-analysis of survivors of pediatric ALL found a high prevalence of obesity in ALL survivors regardless of survivors' receipt of cranial radiation, gender, or age at diagnosis [564]. Another study indicated a relative risk of obesity of 1.97 in association with radiation therapy to the head and neck, cranium, or thorax [4]. However, other investigators found no difference in the prevalence of overweight and obesity among survivors and the general population [565]. In this latter study, an age of younger than 6 years and overweight and obesity at the time of diagnosis were stronger factors than cranial radiation for overweight and obesity in adulthood [565]. Data from a longitudinal analysis of a multiethnic cohort of pediatric ALL survivors suggest that weight gain within the first year after diagnosis is more strongly associated with long-term increased BMI than early (within 30 days) weight gain [566].

Childhood cancer survivors should be weighed yearly at the time of routine history and physical examination and a body mass index should be determined. The patient's height should be considered when reviewing the body mass index, as a high body mass index may be the result of normal weight gain but relative loss of height. Primary care practitioners should emphasize the importance of weight control through healthy eating habits and exercise in preventing the development of such chronic conditions as diabetes, hypertension, dyslipidemia, and cardiovascular disease [319,526,534].

Gonadal Dysfunction

Gonadal dysfunction can occur in both boys and girls as a result of alkylating agents or radiation therapy [273,526,534]. In boys, testicular dysfunction may include delayed or arrested puberty, oligospermia, azoospermia, or infertility. In girls, ovarian dysfunction may include delayed or arrested puberty, premature menopause, or infertility [216,273,319,526,534]. The risk is dose-dependent and is increased when an alkylating agent is used in combination with radiation or an anthracycline [216,534]. The majority of these late effects arise from radiation-induced uterine injury. Such injuries are reported at higher incidence in adult survivors of childhood cancers who were exposed to uterine radiation (i.e., pelvic, spinal, or total-body) during childhood. Findings from long-term follow-up studies of the reproductive performance of female survivors provides some reassurance to them by documenting that pregnancy and live birth rates were not significantly compromised in survivors, including those who had been treated with alkylating agents and had not received pelvic, cranial, and total-body irradiation [567]. The relative risk for male infertility has been reported to be 9.03 in association with use of an alkylating agent and 10.6 in association with both an alkylating agent and an anthracycline [4]. Cranial radiation is correlated with an increased risk for premature puberty, especially for girls treated before 8 years of age [273]. The risk of infertility is low and is directly related to gonadal exposure to radiation and age at the time of treatment [568,569]. The possibility of maintaining gonadal function in the presence of higher cumulative doses of alkylating agents is greater in girls than for boys [534].

The annual history and physical for survivors at high risk for gonadal dysfunction should include evaluation of the puberty stage and pace, sexual function, and Tanner assessment [534]. Baseline levels of follicle-stimulating hormone, luteinizing hormone, and estradiol (girls) or testosterone (boys) should be determined at 13 years of age for girls or 14 years of age for boys and as clinically indicated [534].

Neurocognitive impairment is among the most common and distressing late effect of pediatric cancer treatment [4,319,457,526,570,571,572]. In one study, the relative risk of severe cognitive dysfunction among more than 10,000 survivors was 10.5 compared with sibling controls [3]. In another study, cognitive or psychosocial disorders represented approximately 8% of all severe, life-threatening, or disabling events [4]. A cancer experience alone increases the risk for neurocognitive impairment, and the interruption of regular schooling during treatment can have serious effects on academic performance [534]. Cranial radiation is the greatest contributor to the risk for neurocognitive deficits. The highest risk is associated with cranial radiation in combination with an age of younger than 3 years at the time of treatment or female sex [534]. The intensity of treatment and the time since treatment are also factors [534]. The replacement of prophylactic cranial radiation with intrathecal chemotherapy has led to a decrease in neurocognitive toxicity, but intrathecal chemotherapy (primarily methotrexate) is associated with some level of deficit in approximately 30% of survivors [570,571,572].

A variety of impairments have been reported, and the most common deficits are related to [56,319,471,534,571]:

The deficits in survivors of leukemia and lymphoma are more often related to information processing [534].

The findings of studies on the academic status of survivors support the high reported rate of neurocognitive impairment. In a Childhood Cancer Survivor Study involving 12,430 survivors and 3,410 siblings, significantly more survivors used special education services (23% compared with 8% for siblings) [572]. The results indicated that the greatest differences were found for survivors who were younger than 6 years of age at the time of treatment or who had a diagnosis of CNS tumors, leukemia, or Hodgkin lymphoma [572]. Survivors of leukemia and non-Hodgkin lymphoma were among those who were least likely to finish high school [572]. In addition, a Canadian survey of parents of childhood cancer survivors showed that significantly more survivors than controls had educational or other school problems (46% compared with 23%), had repeated a grade (21% compared with 9%), had attended a learning disability program (19% compared with 7%), or had participated in a special education program (20% compared with 8%) [457]. Leukemia survivors were among the subgroups that were more likely to have educational problems. While there is longstanding recognition that these effects span the whole illness trajectory and continue beyond treatment completion, further clarity is needed on the specific barriers and facilitators to education during cancer treatment and beyond, as well as on the experiences of children and adolescents across the full range of education settings (e.g., hospital, home, virtual, original school of enrollment), in order to determine which interventions are successful in improving access and experience from their perspective [573].

The COG long-term follow-up guidelines recommend that the annual examination for all childhood cancer survivors include an assessment of educational and/or vocational progress [534]. A baseline formal neuropsychologic evaluation is recommended, with follow-up testing for survivors who demonstrate evidence of impaired progress. Primary care practitioners should alert parents to signs of neurocognitive impairment, emphasizing the importance of evaluating any learning-related problems [56]. In addition, primary care practitioners should refer patients with neurocognitive deficits to a school liaison or community service to facilitate access to educational services or vocational rehabilitation.

Dental abnormalities have been associated primarily with radiation to the head and neck area or brain and also with chemotherapy. Abnormalities include tooth agenesis, enamel dysplasia, blunting of roots, incomplete calcification, high plaque index, arrested tooth development, and tooth discoloration [216,273,319,526]. The maturity of the teeth appear to be a substantial factor, as the most severe abnormalities have been found in children who are younger than 5 years of age at the time of treatment [273]. Dental caries and periodontal disease are also associated with bone marrow transplantation with chronic graft-versus-host disease [534].

Despite the prevalence of dental abnormalities, the rate of routine dentist visits among survivors of childhood cancer is lower than that recommended for the general population [574]. Primary care practitioners should encourage patients and parents to maintain regularly scheduled dentist visits, with dental cleaning every six months and an oral examination yearly [534].

Psychosocial disorders are thought to be related to physical function, well-being, and anxiety related to future health needs [444]. As such, survivors of childhood cancers would be vulnerable to such disorders. However, long-term studies have shown contrasting results with respect to the psychological health of childhood cancer survivors. In general, the psychological health of survivors is similar to that of the general population, but a small percentage of survivors and family members have difficulty with coping and with personal and social skills [460].

A meta-analysis of 20 studies showed no deficits in measures of anxiety, depression, and self-esteem when results for survivors were compared with those for population norms or matched controls [444]. Following that study, the CCSS reported on the psychological outcomes for survivors of pediatric leukemia and lymphoma [445]. That study indicated that survivors were approximately 1.7 times more likely to report symptomatic levels of depression or somatic distress compared with sibling controls, but the rate of depression and distress were similar to those among the general population [445,575].

In another Childhood Cancer Survivor Study, the percentage of leukemia and lymphoma survivors who reported an adverse effect on mental health or anxiety was similar to the overall population of survivors. One exception was a greater likelihood for Hodgkin lymphoma survivors to report cancer-related anxiety [537]. The authors of the report suggested that the older age at the time of Hodgkin lymphoma diagnosis may account for this difference, as the patient is better able to understand the seriousness of the cancer diagnosis [537]. This rationale supports the finding of post-traumatic stress disorder symptoms in approximately 20% of adolescent survivors [576].

Age and time since diagnosis may be factors in reported levels of post-traumatic stress disorder [577]. In one study, 255 children with cancer, stratified by time since diagnosis, and 101 demographically matched peers were assessed for post-traumatic stress disorder using structured interviews by both child and parent reports and survey measures of post-traumatic stress symptoms by child report. Cancer was identified as a traumatic event by 52.6% of children with cancer, declining to 23.8% in those who were five years or more from diagnosis [577].

The authors of one study sought to estimate the prevalence of emotional distress in a large cohort of adult survivors of childhood cancer and to evaluate the inter-relationship of risk factors, including cancer-related late effects. A total of 1,863 adult survivors of childhood cancer (median age at follow-up: 32 years) completed comprehensive medical evaluations. Elevated global distress was reported by 15.1% of survivors, and childhood cancer-related morbidities, including pain and learning or memory problems, appear to be directly and indirectly associated with elevated distress symptoms decades after treatment [578].

Sociodemographic factors related to psychosocial disorders among childhood cancer survivors are similar to those of the general population, with higher rates associated with female sex, lower level of educational attainment, and lower socioeconomic status [445,537]. One systematic review assessed unemployment in adult childhood cancer survivors compared with the general population, and its predictors [579]. Overall, approximately one in six survivors was unemployed and survivors were more likely to be unemployed than controls. Survivors of brain tumors were more likely to be unemployed. Younger age, female sex, radiotherapy, and physical late effects were predictors of unemployment [579]. Because the effects of prolonged psychological distress can be severe, assessment for depression, anxiety, post-traumatic stress, and social withdrawal should be carried out annually [534]. Special attention should be directed toward female survivors, Hispanic survivors, and survivors with low socioeconomic status. Primary care practitioners should refer patients with emotional difficulties for psychological consultation.

The CCSS cohort was used to compare the incidence rates of infections among five-year survivors of pediatric cancer with the rates of siblings [580]. Compared with that of the U.S. population, survivors were at an increased risk of death from infectious causes, with girls/women and those exposed to total body radiation being at greatest risk. Survivors also reported higher rates of overall infectious complications than siblings [580]. As part of the Swiss CCSS, a questionnaire was sent to all registered survivors (20 to 40 years of age), all aged younger than 16 years at diagnosis who had survived five or more years [581]. Fewer survivors than siblings had life partners, and fewer survivors were married than siblings and participants in the Swiss Health Survey. Older and female survivors were more likely to have life partners. Survivors who had undergone radiotherapy, bone marrow transplants, or who had a CNS diagnosis were less likely to have life partners [581].

It is imperative that every childhood cancer survivor have written documentation of his or her specific diagnosis and all diagnostic and treatment details [534]. The concept of survivorship care plans has been developed as a way to facilitate better follow-up care for all survivors [584,585]. Among the basic elements of a survivorship care plan are [534,585]:

One study suggests, however, that a majority of cancer survivors and their physicians have not yet embraced survivor care plans [586]. From April to July 2012, researchers conducted in-depth telephone interviews with 53 adult survivors (average age: 39.1 years) recruited from the Utah Cancer SEER Registry. Participants were randomly selected from sex, age, and rural/urban strata and were younger than 21 years of age at the time of diagnosis. Participants were asked if they had a primary care physician, if they discussed their cancer history with their physician, and if they had interest in a survivor care plan. Most (83%) survivors had a current primary care physician; nearly 50% were not concerned about their health despite having had cancer; and few had a follow-up care plan, although more than half felt that a survivor care plan could empower their medical decision making. However, more than 30% of survivors were skeptical about the usefulness of a survivor care plan and some expressed concern about healthcare costs [586].

The National Children's Cancer Society has introduced a new online Late Effects After Treatment Tool (LEATT), which is available online at https://leatt.thenccs.org. The tool was designed to help survivors learn about potential late effects of childhood cancer. The LEATT provides personalized results, based on the information that the survivor enters, and it takes less than 10 minutes to complete. The results describe potential late effects and symptoms to look for, provide recommendations for follow-up care, and offer preventive tips to help reduce the risk of further health issues [587].

A valuable resource for patient and family education is Health Links, a series of educational handouts based on the COG guidelines for long-term care of childhood cancer survivors. These handouts are designed to enhance patient follow-up visits and broaden the application of guidelines. The entire set of Health Links is available online [588].The handouts address more than 40 topics within the guidelines, and each handout contains a brief overall description of a specific late effect, as well as [533,534]:

The health behaviors of childhood cancer survivors have not been well-studied, but overall, young cancer survivors report lifestyle behaviors that are at least as healthy as the age-matched population [56,532]. Behaviors have been shown to vary according to race/ethnicity; overall, Black survivors have reported better preventive practices and have been less likely to engage in risky behaviors such as smoking and alcohol consumption [589]. In one study, 40% of survivors reported eating a nutritious diet, 64% said they practiced sun protection, and 94% said they did not smoke [590]. In a later and much larger study, the rate of physical inactivity was high (52% to 72%), with the lowest rate for Hispanic male survivors and the highest rate for Black female survivors [589]. In that same study, the rates of smoking among Black and Hispanic survivors were lower than the rates among their peers in the general population [589].

In a case-control study published in 2014, researchers sought to identify factors other than cancer-related treatment and presence/severity of chronic health conditions that might be associated with late mortality risk among adult survivors of pediatric cancer [591]. A total of 445 participants who died from causes other than cancer recurrence/progression or non-health-related events were compared with 7,162 surviving participants matched for primary diagnosis, age at baseline questionnaire, time from diagnosis to baseline questionnaire, and time at risk. Exercising fewer than three days per week, being underweight, increased medical care use, and self-reported fair-to-poor health were associated with an increased risk for all-cause mortality [591]. A report from the Childhood Cancer Survivor Study found that poor physical activity during adolescence, parents with less than a college education, previous treatment with cranial radiation, limitations of activity due to health or mobility restriction, and low self-esteem were associated with level of psychosocial functioning among childhood and young adult cancer survivors [592].

Primary care practitioners should make sure that childhood cancer survivors understand that the risks associated with unhealthy lifestyle behaviors may be increased because of their previous cancer treatment. In addition, clinicians should ensure that survivors know the importance of a healthy diet, regular exercise, routine sun protection, avoidance of tobacco, and moderation in alcohol consumption. Survivors should also be educated about the signs and symptoms of depression and of ways to alleviate stress and anxiety.

The cancer screening practices and compliance with healthcare recommendations have been less than optimal among childhood cancer survivors, and most survivors have contact with primary care practitioners rather than oncology specialists or follow-up clinics [532,550,569,584]. These findings indicate the importance of primary care practitioners assuming responsibility for the long-term follow-up care of childhood cancer survivors; however, barriers to doing so remain. In one study, a minority of primary care physicians perceived that their medical training was adequate to recognize late effects of chemotherapy (27.6%), cancer surgery (36.6%), and radiation therapy (38.1%). Most (93%) had never used the COG guidelines, but 86% would follow their recommendations. Most (84% to 86%) stated that they had never received a cancer treatment summary or survivorship care plan, but more than 90% thought these documents would be useful [593]. Although most general internists report involvement in the care of pediatric cancer survivors, many are unfamiliar with surveillance guidelines and uncomfortable caring for survivors, citing access to guidelines and treatment summaries as the most useful resources in caring for survivors and preferring to care for survivors in collaboration with a cancer care center [594]. A survey of U.S. and Canadian family physicians involved in the care of pediatric cancer survivors revealed similar attitudes. Clinicians also cited surveillance recommendations and clinical care guidelines as being the most useful in assisting them with survivor care and were willing to care for survivors in consultation with a cancer care center [595].

Due to their heightened risk of developing late-occurring adverse outcomes, pediatric cancer survivors are advised to receive follow-up care in specialized survivor clinics. Research indicates that most survivors (71%) who do not attend a specialized clinic were "not aware" of their availability [596]. Clinicians should use every opportunity (e.g., acute care visits, parental or sibling visits, annual examinations) to emphasize the importance of long-term follow-up care in a survivor care clinic and adherence to recommended screening guidelines for survivors.