Aims-To determine the role of polymerase chain reaction (PCR) based minimal residual disease (MRD) detection of leukaemia specific DNA in testicular relapse in childhood acute lymphoblastic leukaemia. Methods-DNA was obtained from archival testicular and bone marrow samples from boys with acute lymphoblastic leukaemia who relapsed in the testes. Overlapping DJ H clone specific primers derived from clonal immunoglobulin heavy chain (IgH) gene rearrangement in each case were used to analyse testicular or bone marrow DNA. Results-Histologically normal end of treatment testicular biopsies in the five patients in longterm remission were all MRD negative, but MRD positive in three of six boys with subsequent testicular relapse. Histologically normal bone marrow samples taken at the end of treatment were MRD negative in five of seven cases, but MRD positive in all cases at the time of isolated testicular relapse. Three boys with unilateral testicular relapse underwent unilateral orchidectomy, rather than bilateral testicular irradiation, as part of their treatment. Two of these boys were MRD positive in the histologically uninvolved testes, and both had subsequent relapses either in the testes or the bone marrow, while the MRD negative patient has not had a testicular relapse. Conclusions-The presence of MRD in testicular tissue can be assayed with a PCR based method to detect clone specific antigen receptor gene rearrangements. In this setting, PCR is more sensitive than conventional testicular histology for predicting clinical outcomes. MRD assays might be useful in the management of boys at the time of isolated testicular relapse, to confirm the presence of unilateral testicular disease.

e20716 Background: Febrile neutropenia (FN) is a major complication of chemotherapy, costly in terms of morbidity, mortality and associated financial expenditure. The present study was conducted with the goal of highlighting FN as a serious problem in Pakistan, with the longer term objective of improved cancer survival, reduction in length of stay (LOS) in hospital, morbidity, mortality and costs in our existing developing country scenario. Methods: A cross-sectional descriptive study was conducted on patients, >18 years, admitted with FN as a consequence of chemotherapy at a referral hospital in Karachi from 1st September 2006 to 30th April 2007. Results: A total of 80 patients [43 (53.8%) males and 37 (46.2%) females] were selected. The mean age was 47.4 (SD ±16.6; range 18–79) years. Sixty eight patients (86%) were < 65 years, 50% were < 50 years. Overall, in hospital mortality was 11%; 4% for patients on granulocyte colony stimulating factor (G-CSF) prophylaxis as against 20% for those without. The cause of death was either pneumonia or septic shock. Mean LOS was 7.53 (SD ±3.8; range 2–17) days. Hematological malignancies, older age, severity of dehydration, pneumonia and culture positivity were significantly associated with LOS and death. Those above 50 years of age were 1.5 times as likely to be hospitalized longer and > three times as likely to die. Bacteremia conferred a 5-fold and pneumonia an 8-fold increase in the risk of death. Conclusions: The results of this study indicate that age, vital instability, dehydration, high creatinine, culture positivity and hematological malignancies are high risk factors in chemotherapy induced FN. Identification of FN risk factors with poor outcomes may help in devising protocols for modified dosage or including GCFs initially. This may help reduce the cost of cancer care as well as mortality and morbidity. Prospective studies of FN in multiple centers in Pakistan may be beneficial in evaluating these risk factors further. No significant financial relationships to disclose.

No abstract