Screening for neuroblastoma: 20 years and still no answer.

Abstract

IF PREVENTION is not possible, then early diagnosis must be the goal in any form of cancer. There is now ample evidence that screening for some adult cancers is worthwhile, cost-effective and saves lives [l, 21, but, to date, the same is not true for children. The lives of many children with cancer are saved, but at the cost of expensive multimodality therapy and with the ‘late effects’ of the treatment, including an increased risk of second malignancy [3]. The opportunities for screening for childhood cancer in the population are limited, for a variety of reasons. Firstly, the relative rarity of cancer in the young (only one child in 1800 develops leukaemia, the most common cancer of childhood before the age of 15 years) makes screening appear expensive per case detected, although this cost might be counterbalanced by the large number of potential life years saved by an effective programme. Secondly, the relative effectiveness of current therapy, with an overall survival rate in excess of 65-70%, argues against screening for most childhood cancers. For example, screening might be possible for Wilms’ tumour using serial abdominal ultrasound examinations, but the high ‘cure’ rate @O-85%) with current treatment would suggest that even this would not be worthwhile. Neuroblastoma is the one childhood cancer for which screening might be beneficial. The prognosis for disseminated (stage 4) neuroblastoma diagnosed in clinical practice has only improved slightly over the last 20 years, and where progress has been made, it has included the use of very intensive chemotherapy, often with autologous or allogeneic bone marrow rescue [4]. The outlook for children diagnosed with localised disease or those diagnosed at a young age, particularly those under the age of 1 year, is very much better than for older children or those with stage 4 disease [5]. In addition, at least 85% of children diagnosed as having neuroblastoma have raised urine levels of catecholamine metabolites, detectable by simple biochemical tests on spot urine samples, and which could be undertaken on whole populations of babies at an early age [6, 71. The ease of the collection of urine from filter papers placed on wet nappies seems to complete the ‘setting of the scene’ for an ideal