Incidence

ALL is the most common hematologic malignancy of childhood. The majority of incidence data are generated by high-income countries (HICs); in the United States, there are approximately 42 cases of ALL per 1 million children. However, less than 15% of the population in Africa and Southeast Asia is captured by established cancer registries and it is thus very difficult to know if these numbers are accurate on a global scale. Historically, the incidence has seemed to be lower in less-developed countries with a mean annual incidence of 16.5 cases per million population. However, the incidence does seem to be steadily increasing at a greater rate than in HICs, likely owing to a combination of improved awareness, better reporting, and decreases in other contributors to childhood mortality.

Treatment and outcomes

ALL can be cured in almost 90% of children in HICs using multiagent systemic and intrathecal chemotherapy for a total duration of therapy of 2 to 3 years. Induction, the first phase of treatment, consists of more intensive therapy and patients often require prolonged hospitalization and systemic antimicrobials. Repeat examination of the marrow occurs at the end of induction to assess for early response and provide intensification of therapy in patients not in remission. High-risk patients receive a second induction course and a minority of children, primarily those with central nervous system involvement, will require cranial radiation therapy. Maintenance therapy is of much less intensity and occurs primarily in the outpatient setting. However, the care remains complex. Maintenance involves a rotating combination of oral, intravenous, and possibly intramuscular drugs, of which some are administered daily, some weekly, and some monthly. Repetitive visits to the cancer center are required.

Approaches to treatment and outcome have been quite variable in the LMIC setting; in general, as the intensity of therapy increases the number of relapses has decreased, but treatment-related mortality has increased. Thus, attempts have been made to match therapy intensity to the level of supportive care available, as well as to use risk stratification to deliver the most intense therapy to the patients with the worst prognosis based on presenting features and early response to therapy. All approaches have used an induction phase and a maintenance phase with intermittent intrathecal chemotherapy administration.

Given that there are still reported differences in outcomes related to disparities in socioeconomic status in HICs, the challenges of improving outcomes in the LMICs is enormous. There is, unsurprisingly, a wide range reported for disease-free survival (DFS), from as low as 15% to greater than 60%, depending on the patient population and resources available. Standard factors used for risk stratification—namely, age, initial white blood count, and prednisone response—have been assumed to apply globally, although a large study from India found that age had no impact on outcome and that risk factors were center specific even with the use of one standardized protocol within 1 country. There are many potential factors contributing to inferior survival with some more amenable to intervention than others:

• 1.
  

  Innate biological differences: TEL-AML1 is the most common acquired genetic abnormality in pediatric B precursor ALL, occurring in 20% to 30% of patients in the United States. Patients harboring this transcript are known to have an improved DFS. Although the impact on prognosis seems to be immutable, the proportion of TEL-AML1–positive patients may vary by region and occur less frequently in some areas of the world, such as south Asia. Similarly, a higher percentage of patients are reported to have T-cell versus precursor B ALL in Egypt and rural India. This biologic variability results in an increased incidence of high-risk features in some populations and thus would contribute to poorer outcomes for the group as a whole.

  
  
• 2.
  

  Delay in diagnosis: Delay between development of symptoms and initiation of therapy can impact survival from solid tumors at risk for metastasis. This may not be as problematic in hematologic malignancies, which are systemic at the time of presentation. In addition ALL may be recognized and therapy sought more quickly than in other childhood malignancies such as Hodgkin lymphoma (HL), even in resource-poor settings. A separate issue is that some children with ALL may receive pretreatment with steroids for symptoms thought to reflect other diagnoses, such as arthritis, idiopathic thrombocytopenic purpura, or simply failure to thrive. Monotherapy with prednisone could negatively impact DFS owing the development of steroid-resistant clones before the initiation of ALL-directed therapy.

  
  
• 3.
  

  Undernutrition is prevalent among children in LMICs, and especially so in children with cancer at diagnosis. In a large sample in Central America approximately two-thirds of such children were classified as severely malnourished and had a notably lower survival than those who were not so deprived nutritionally. However. It was demonstrated subsequently, in a cohort with ALL in Guatemala, that restoration of normal nutritional status within 6 months of the initiation of treatment was associated with a survival rate similar to that in children who had maintained a normal nutritional status throughout.

  
  
• 4.
  

  Inadequate supportive care measures: In the United States, treatment-related mortality is less than 5% in children with ALL. This number is significantly higher in resource-poor settings and can account for up to one-fourth of deaths. The main causes of mortality are infection and bleeding. Provision of better antibiotic therapy and more rapid access to transfusion support are thus needed. However, improved education of providers, written supportive care guidelines, and the identification and treatment of malnutrition and concurrent infection may also decrease the proportion of deaths owing to treatment-related complications.

  
  
• 5.
  

  Delays in therapy: There are many reasons for delays in therapy, including missed appointments, patient-specific medical complications (fever, prolonged neutropenia), and drug unavailability. One study of ALL in Colombia found that delays between starting induction and beginning maintenance therapy (a 24-week period according to protocol), regardless of whether owing to medical issues or other reasons, was associated with a significantly inferior DFS.

  
  
• 6.
  

  Abandonment: There have been many attempts to address the intertwined issues of refusal (refusing treatment), noncompliance (lack of cooperation with recommended therapy), and abandonment (termination of therapy prematurely). Key components of successful intervention strategies include family education about therapy duration, side effects, and the potential curability of cancer in addition to increased social support systems including housing, transportation, and stipends to cover loss of income.

Core resources required to treat acute lymphoblastic leukemia in low- and middle-income countries

• •
  

  Adequate pathology capacity that allows for rapid diagnosis, as well as determination of immunophenotype (B- vs T-cell disease) and other features necessary for risk stratification at diagnosis and at the end of induction therapy.

  
  
• •
  

  Access to a safe and reliable supply of red cells and platelets for transfusional support, particularly during induction.

  
  
• •
  

  Ability to adequately diagnose and treat bacterial, fungal, and viral infections during periods of prolonged neutropenia.

  
  
• •
  

  Adequate family supports (transportation, outpatient housing near the hospital, community health worker involvement) so that maintenance chemotherapy can be administered consistently.

Future directions

Incidence

Acute myelogenous leukemia (AML) accounts for approximately 15% of acute pediatric leukemia in HICs. In comparison with ALL, there seems to be no difference in the incidence between HICs and LMICs.

Treatment and outcomes

There is much less published data on treating AML versus ALL in LMICs. Compared with ALL, therapy is of much shorter duration but of correspondingly greater intensity. The overall duration of therapy is 4 to 6 months, and the majority of that time is spent in the hospital receiving systemic and intrathecal chemotherapy and intensive blood product and antimicrobial support and use of granulocyte colony-stimulating factor is routine. Radiation therapy is not used routinely. Reports are scarce from LMIC settings, although it seems that between 0% and 40% of children are long-term survivors. Treatment related deaths are reported to occur in more than 20% of patients, reflecting the problems encountered in HICs before the routine use of antimicrobial and antifungal prophylaxis. Predicting those at greatest risk of treatment-related mortality remains problematic, although it is assumed that comorbidities such as malnutrition are important and this remains an area requiring more investigation. Bone marrow transplantation is the therapy of choice for refractory or recurrent AML in HICs; in the resource-limited setting, treatment may be best directed toward palliation. Children with late relapses (>18 months from diagnosis) are a possible exception because they can experience a prolonged period of remission with salvage chemotherapy; cure, however, remains very unlikely.

Core resources required to treat acute myelogenous leukemia in low- and middle-income countries

• •
  

  Adequate pathology capacity that allows for rapid diagnosis, as well as identification of other features (FLT3 positivity, monosomy 7) necessary for risk stratification at diagnosis and at the end of induction therapy.

  
  
• •
  

  Ability to manage leukostasis, bleeding, and tumor lysis syndrome that can occur at presentation and during induction therapy.

  
  
• •
  

  Access to a safe and reliable supply of red cells and platelets for transfusional support.

  
  
• •
  

  Availability of growth factors as well as broad spectrum antimicrobials for prophylaxis and treatment of infection with Gram-positive organisms and fungus.

Future directions

Standard therapy for CML treated with TKIs involves routine assessment of quantitative levels of bcr-abl from the peripheral blood throughout the course of treatment. Patients who have evidence of early disease progression are switched to another TKI. Currently, few countries have the infrastructure to monitor patients consistently in this fashion and even fewer have access to the expensive subsequent generation TKIs.

Incidence

HL seems to occur more commonly in LMICs perhaps owing to the high incidence of Epstein–Barr virus (EBV) infection in the population. Latent EBV can be found in Reed–Sternberg cells, the malignant cell type in HL, in 50% of patients in LMICs, although it does not seem to impact on prognosis. HL also occurs at a younger age in this setting with a peak incidence of 5 to 9 years, which also may be related to the role of EBV infection in this disease.

Treatment and outcome

Accurate staging is critical because it directs both the components and duration of therapy. Evaluation of chest, neck, abdomen and pelvis with computed tomography is the minimum assessment needed. It remains difficult to distinguish normal from pathologic nodes without functional imaging such as F-labeled 2-deoxyglucose PET. This imaging modality is rarely available in resource-poor settings and as a result patients are at risk for being assigned a stage that is either higher or lower than that reflected by the actual burden of disease. Most children present with stage III or IV disease in LMICs. Treatment has been adapted from HICs and most often involves 4 to 6 cycles of combined chemotherapy that can usually be administered in the outpatient setting. Most approaches include radiation therapy to sites of previous disease. Radiation almost always occurs at a different facility and often in a different city or country. Difficulties in coordination of care across settings is always a challenge and can negatively impact outcome as a result of many factors, including treatment delays, loss of disease control, and abandonment of therapy. However, in general, HL is a very treatable malignancy and using a variety of standard published approaches reported DFS is between 60% and 90%, in many settings approaching that reported in HICs.

Incidence

Lymphoblastic lymphoma can be of either a pre-B or T-cell phenotype and is treated identically to ALL.

BL is the most common subtype of pediatric non-Hodgkin lymphoma and is particularly prevalent in sub-Saharan Africa, where it accounts for almost 50% of all childhood cancers. This is thought to be related to endemic EBV infection because more than 90% of cases contain the ebv genome in tumor tissue. Malaria transmission may be an additional factor. T-cell control of EBV is impaired during attacks of malaria and Plasmodium falciparum may also interact directly with B-cell receptors stimulating proliferation of EBV-infected lymphocytes. The incidence varies by region but can be as high as 22 per 100,000 children and account for up to 70% of all new childhood cancers in parts of Tanzania. It does not seem that human immunodeficiency virus has made an impact on the incidence of the disease, although patients with BL coinfected with human immunodeficiency virus seem to have a worse survival. In the African setting, the disease presents most often as a jaw mass, although a recent report suggests that obtaining abdominal ultrasound in addition to a physical examination improves staging accuracy and the incidence of abdominal involvement then becomes similar to that in HIC. The incidence of B-cell large cell lymphoma is not described in LMICs and can be difficult to distinguish clinically and pathologically from BL; these patients can be treated effectively on BL protocols.

Treatment and outcomes

In HICs, B-cell lymphomas can be cured in 90% of children but therapy is very myelosuppressive and requires a prolonged hospital stay and availability of a broad range of antimicrobial agents. Historically, oral cytoxan (cyclophosphamide) alone has been reported to cure a small number of cases of BL and low-dose combination chemotherapy regimens up to one-third of children. Currently in LMICs, risk stratification has improved survival to greater than 50%. The least intense approach for those with low risk disease, as assessed by stage and clinical response to initial therapy, consists only of cytoxan and intrathecal (IT) therapy. Intravenous methotrexate is added for higher risk patients and total duration of therapy is less than 3 months for all groups. Unfortunately, the time to diagnosis is often delayed in resource-poor settings, particularly when children present with findings such as cervical adenopathy, which usually reflects a benign self-limited process. Thus, the vast majority present with nonlocalized disease and require the more aggressive regimen. A consortium of 8 African countries have adopted modified protocols used in HIC by lowering the dosages of drugs most associated with profound myelosuppression. This resulted in an overall survival of 84%, including a 76% overall survival for stage II patients. Of import, treatment-related mortality decreased by 50% (from 26% to 12%) over the 3-year study period, likely related to quality improvement initiatives including regular meetings and review of adverse events. Relapsed/resistant disease remains problematic, although in a small series of patients, retreatment with cytoxan and IT therapy and the addition of vincristine provided disease control in more than one-third of those treated. Patients whose relapse occurred after more than 6 months off therapy and who presented with limited disease were most likely to respond.

Core resources required to treat B-cell lymphoma in low- and middle income countries

Management of tumor lysis syndrome: Tumor lysis syndrome is a frequent problem in patients with advanced stage BL and can lead to severe electrolyte abnormalities, renal failure, and death. Frequent clinical and laboratory monitoring is essential but difficult to ensure in the setting of limited staffing. The use of allopurinol, intravenous hydration, and involvement of caretakers in monitoring urine output has been suggested as a feasible approach to the management of this complication.

Future directions

Improvements in supportive care are key to better survival. Currently almost one-third of those on protocols of moderate intensity succumb to treatment related complications. As in AML, broad spectrum antibiotics and antifungal drugs must be available in addition to growth factor support.

Rituximab, a monoclonal antibody directed against the CD20 antigen found on mature B cells, has been incorporated into many B-cell lymphoma regimens in HICs. It has been shown repeatedly to improve survival in adults but definitive evidence assessing its role in pediatric B-cell lymphomas is lacking. It is expensive and requires close monitoring during administration but may allow for incremental improvement in DFS when combined with the less intense regimens often used in LMICs for these diseases.