G. Wessels

The 2000 Burkitt lymphoma trial in Malawi.

We previously reported 57% 12‐month event free survival (EFS) in Malawian children with stage I to III Burkitt lymphoma (BL) with an intermediate dose chemotherapy protocol lasting 77 days. This protocol was shortened to 42 days and evaluated in children with stage I to IV disease for EFS and toxicity.

Hepatocellular carcinoma and liver tumors in South African children: a case for increased prevalence.

The high regional incidence of hepatocellular carcinoma (HCC) in South Africa also may be present in children of the region, although the link to hepatitis B (HBV) appears less clear. The objective of this study was to assess the incidence and probable causes of HCC in South African children.

Infection with human immunodeficiency virus-1 (HIV) among children with cancer in South Africa†

HIV increases the risk of certain cancers known, or thought, to have an underlying infectious etiology; the impact on the risk of cancer in children is less clear.

Tuberculosis in oncology patients

Abstract Background: There is a dearth of studies addressing the incidence and clinical presentation of tuberculosis (TB) in children with cancer. Aim: To evaluate the incidence of TB in paediatric oncology patients at Tygerberg Hospital, located in a Cape Town area of high TB prevalence, and to describe the clinical characteristics of the disease in this particular group of patients whose treatment typically suppresses their immune response. Methods: We reviewed the records of 625 paediatric oncology patients admitted from 1 January 1991 to 31 December 2005. Of these, 87 received treatment for TB; however, only 57 cases had sufficient data to support a diagnosis of TB and only these were used for further analysis. Results: In the children with TB, acute lymphoblastic leukaemia (ALL) was the most common malignancy (13/57, 22.8%). The incidence of TB in the study group was 9117/100,000/year, which is 22 times higher than the overall TB incidence reported in children from a similar background. Importantly, 47% of the active infections appeared in the 1st 5 months of chemotherapy, suggesting reactivation of latent TB. Conclusions: Identifying latent TB in our patients and providing prophylactic treatment during the initial months of chemotherapy might have prevented disease progression in these cases. Routine screening of paediatric oncology patients for latent TB infection and exclusion of active disease prior to initiation of cancer therapy might be indicated in TB-endemic areas.

High-dose intense chemotherapy in South African children with B-cell lymphoma: Morbidity, supportive measures, and outcome

BACKGROUND Twenty-five percent of South African children aged 6-71 months are undernourished and have stunted growth. The tolerance and efficacy of short, high-dose intense chemotherapy for B-cell lymphomas in such a population were unknown. PROCEDURE Nineteen consecutive children diagnosed with B-cell lymphoma after 1993 at Tygerberg Hospital (TBH) in the Republic of South Africa (RSA) were treated according to the LMB-89 protocol. RESULTS Among the 19 children treated according to the LMB-89 protocol, there were 3 children in group A (completely resected St. Jude stage I and abdominal stage II), 14 in group B (nonresected stage I, nonabdominal stage II, all stage III, stage IV with bone marrow involvement but <70% Burkitt cells and without CNS involvement) and 2 in group C (patients with >70% Burkitt cells in bone marrow and/or CNS involvement). Overall survival for these children was 79% (median follow-up 53.5 months,range 20-70 months) compared to 25% (median follow-up 131 months, range 71-173months) for 24 children who had been treated with COM+/-P prior to 1993 (P = 0.002). Toxicity was noteworthy in the children treated with LMB-89. They had a mean of 2.6 episodes of febrile neutropenia and 1.9 episodes of stomatitis per patient and required intensive support, but there were no toxic deaths. CONCLUSIONS A major step forward was achieved for South African children with B-cell lymphoma. Despite a high prevalence of malnutrition and endemic infections in the RSA, the implementation of the LMB-89 protocol significantly improved survival with manageable morbidity. Our findings suggest that treatment centres that cannot measure methotrexate (MTX) serum levels should not exceed 3.0 g/m(2) of MTX. If supportive care facilities are limited, consideration should be given to reducing the doses of cyclophosphamide and of doxorubicin in the treatment schedules.

Simultaneous Occurrence of Viral-Associated Hemophagocytic Syndrome and Langerhans Cell Histiocytosis: A Case Report

Langerhans cell histiocytosis (LCH) is a class I histiocytosis characterized by the presence of the pathologic Langerhans cell, an unique histiocyte. In contrast to LCH, class II histiocytosis is characterized by the proliferation of mononuclear phagocytes other than Langerhans cells and includes sinus histiocytosis with massive lymphadenopathy, viral-associated hemophagocytic syndrome, and familial hemophagocytic lymphohistiocytosis. Until now, these two classes have been considered separate, if related, entities. We report a 10-month-old girl who presented with pyrexia, hepatosplenomegaly, an eczematous skin rash, anemia, thrombocytopenia, and a markedly elevated serum IgG and IgM antibody level to cytomegalovirus. Histologic proof of both hemophagocytosis in the liver and bone marrow and LCH in the skin was obtained at presentation. The clinical course and response to treatment over 6.5 years is recorded. Although the etiology of both class I and class II histiocytosis remains unknown, we speculate that the monocytic/macrophage disorder, as well as the LCH, were both triggered by virus or viral-related monokines secreted by activated macrophages.

An interinstitutional review of the value of FNAB in pediatric oncology in resource‐limited countries

Fine‐needle aspiration biopsy (FNAB) has been widely accepted as a reliable diagnostic modality in the general pediatric population, but its role in pediatric oncology still remains elusive. With new treatment protocols subscribing to preoperative chemotherapy, the need for a quick, minimally invasive, and accurate diagnostic procedure has arisen. This study assesses the feasibility of FNAB in childhood malignancies to render a specific diagnosis on which treatment can be initiated. An 11‐year retrospective study was done on FNABs in patients 19 years and under referred for clinically malignant mass lesions. Cases were confirmed with histology, immunocytochemistry, flow cytometry, or clinical follow‐up. Of the 357 patients referred for FNABs, 36 patients were lost to follow‐up and 31 FNABS were inadequate. A total of 290 cases were included in the study, of which 68 (23%) cases were benign and 222 (77%) were malignant. The most frequently occurring tumors were nephroblastoma (68), non‐Hodgkins lymphoma (39), rhabdomyosarcoma (22), Hodgkins lymphoma (22), and neuroblastoma (22). The sensitivity of the procedure for neoplasia was 96.6%, the specificity 97.0%, positive predictive value 99.0%, and negative predictive value 90.1%, with a diagnostic accuracy of 96.7%. The ability of FNAB to enable a specific diagnosis to be made, that is correct and accurate subtyping of the tumor on which chemotherapy or radiotherapy could be commenced was 75.7%. This study shows that FNAB can be used with confidence to confirm malignancy in children. With clinicoradiological correlation and the aid of ancillary techniques, FNAB allows a rapid and accurate preoperative diagnosis for definitive therapy commencement in most cases. Diagn. Cytopathol. 2012.

Haematological abnormalities in children with tuberculosis.

Over a 16 month period 307 children with suspected tuberculosis (TB) and an available full blood count (FBC) seen at Tygerberg Hospital in South Africa were evaluated and categorized as confirmed (A), probable (B), and no TB (C) according to WHO criteria. There was no difference in the mean age of the 168 group A (33.6 months), 83 group B (34.4 months), and the 56 group C (31.6 months) children. A lower mean haemoglobin (Hb 10.2 vs. 10.8 g/dl) was the only significantly different haematological parameter in children with TB compared with the comparison group (Group C). There were no differences in median total white cell count, neutrophils, lymphocytes, monocytes, platelets, or the proportion of children in each group with anaemia, microcytosis, neutrophilia, neutropenia, lymphocytosis, lymphopenia, monocytosis, thrombocytosis or thrombocytopenia. The most common haematological abnormalities in children with TB were the presence of anaemia, neutrophilia, and monocytosis but these changes were found with equal frequency in control patients. Although haematological abnormalities are fairly common in children with TB, in a developing country these abnormalities also occur frequently in children with other non-tuberculosis respiratory infections. An FBC has no diagnostic predictive value when investigating a child for TB.

An analysis of prognostic variables in acute lymphocytic leukaemia in a heterogenous South African population

The records of all 96 children below the age of 15 years diagnosed with acute lymphoblastic leukaemia at Tygerberg Hospital in the Republic of South Africa between 1983 and 1995 were reviewed to determine risk factors which may predict poor outcome. Age < 2 and > 8 years, and white cell count > 20 x 10(9)/l at diagnosis were significant predictors of poor outcome. Sex, FAB classification, immunophenotype, hepatomegaly, splenomegaly, BFM risk score, and the presence of mediastinal glands did not predict outcome. The presence of the established risk factors could not adequately explain the difference in 5-year event-free survival in the three ethnic groups which was 67 per cent in white, 17 per cent in black, and 38 per cent in children of mixed ethnic origin. In an attempt to improve survival in black children, our stratification of risk groups will in future be based on factors that include ethnicity, age and WCC > or = 20 x 10(9)/l at diagnosis. Pediatric oncology services in developing countries should adapt therapy to the risk factors of their local populations.

Viral isolates during febrile neutropaenia in children with cancer.

We prospectively studied South African children with cancer for viral isolates during episodes of febrile neutropaenia. Viruses were found in seven (31.8 per cent) and bacteria in five (22.7 per cent) of 22 episodes. The most common isolate was the herpes simplex virus and the most common source was from nasopharyngeal aspirates. There was no dual detection of viral and bacterial isolates. This study emphasizes the important contribution of viruses to febrile neutropaenia.