Incidence of Childhood Cancer in Rural India

Although 70% of the Indian population lives in villages, we have relatively little real information about cancer in them as most of the population-based cancer registries (PBCRs) are urban. The incidence and mortality patterns of these urban PBCRs cannot provide an accurate picture for the entire country as the environmental exposures and the gene-environment interactions leading to cancer could be substantially different within different part of India.

In such a context the first report of the rural PBCR at Dindigul Ambilikkai in Tamil Nadu provides us useful information about cancer incidence (including childhood cancer incidence) in rural India (Ref - Swaminathan et al, Cancer Epidemiology, Nov 2009). The reported incidence of cancer in boys less than 15 years of age of 56 million person years and in girls of 46 million person years is similar to reports from the two other rural PBCRs in India and much less than that reported from the urban PBCRs in India and from other parts of the world (Ref - Arora et al, Indian Journal of Cancer, Oct-Dec 2009).

Under-diagnosis (and consequently under-registration) of childhood cancer due to lack of adequate access to diagnostic and clinical services is likely to be the main reason for the lower incidence of childhood cancer seen in rural India (Ref - Arora, Cancer Epidemiology, Feb 2010) and (Ref - Swaminathan et al, Cancer Epidemiology, Feb 2010)

Enabling Timely Diagnosis of Brain Tumours in Children

Worldwide, tumours of the brain (and other parts of the central nervous system) constitute nearly 25% of all childhood cancer (second only to acute leukaemias). Their incidence in India is lower and this is partly related to underdiagnosis as a result of inadequate neuro-radiological and neuro-surgical facilities (Ref - Arora et al, Indian Journal of Cancer, Oct-Dec 2009)

Presenting features of the brain tumour depend on the age of the child as well as the location of the tumour in the brain. However, they can be varied and include headache, nausea/vomiting, visual changes, motor changes, seizures, altered sensorium and others. The non-specificity of some of these symptoms can lead to a delay in parents seeking medical advice and in physicians reaching a diagnosis. These delays can have negative impact on outcomes.

The Children's Brain Tumour Research Centre in Nottingham University has developed evidence-based guidelines to help clinicians recognise and diagnose brain tumours in children in a timely fashion. These guideline have application across all settings and populations including in the Indian context and deserve to be widely disseminated.

(Ref - Wilne et al, Archives of Diseases in Childhood, 2010) and (Ref - Royal College of Paediatrics and Child Health Guidelines)

Proportion of Translocations in Childhood ALL in Pakistan is similar to that in India

TEL gene rearrangement due to the 12;21 chromosome translocation is the most common molecular genetic abnormality in childhood acute lymphoblastic leukemia (ALL) and is seen in around 25% of children with pre-B cell ALL (studies mainly from Europe and North America). In this translocation the 5’ part of the TEL (ETV6) gene fuses with almost the entire AML1 (CBFA2) gene, producing the chimeric transcript ETV6-CBFA2.

Figure from www.kreatech,com

It is associated with a better prognosis in ALL . In contrast to the West, the frequency of the translocation TEL/AML1 in children in India with ALL is around 7% (Ref - Siraj et al, Leukemia, June 2003). Now in a study of 50 children with ALL from two institutes in Lahore, Pakistan the frequency of the TEL/AML1 translocation was a similar 6% (Ref - Faiz et al, Journal of Pediatric Hematology Oncology, April 2010). Interestingly, the frequency of another translocation which confers a poor prognosis, t(9;22) or BCR/ABL, was 24% and is much higher than that reported from India and elsewhere.

Clearly, these results are from a hospital-based case series and further larger studies in Pakistan would be necessary. The fact that 42 of these 50 ALL children with ALL were boys illustrates the limitation of such hospital-based case series.

Addendum (08/06/2010) - A recent report shows that the frequency of TEL/AML1 in Far East Asia (Japan, Korea, China, Hong Kong, Chinese in Singapore, and Taiwan) based on a pooled analysis of 1321 children with ALL was 13.4% (Ref - Liang et al, Pediatr Blood Cancer, 2010). This is lower than that reported from the West but higher than that from the Indian subcontinent.

Ovarian Germ Cell Tumours in Children

Tumours of the ovaries represent around 4% of tumours overall in females and around 2% of all tumours in children less than 15 year of age. In children and adolescents, germ cell tumours of the ovary are by far the most common pathology (around 70%) among ovarian tumours while in adults, carcinoma of the ovarian epithelium accounts for 85-90% of the ovarian tumours.

Dr Biswajit and his colleagues from a tertiary cancer centre from Chennai have recently published outcomes of 40 girls less than 18 years of age with ovarian germ cell tumours who were managed in their institute from 1990 to 2002 with cisplatin-based chemotherapy and surgery (Ref - Biswajit et al, Journal of Pediatric Hematology Oncology, March 2010). Nearly 2/3^rd of the patients presented in Stage III-IV. Delays in diagnosis and treatment could have been due to patient-related factors or due to healthcare-related delays although this was not specifically studied. The 5 years disease-free survival was 72.8% (mainly because of relapses in 25% of patients) and overall survival was 94.9%. Similar results were reported from Tata Memorial Hospital in Mumbai nearly 15 years ago although they had less relapses (Ref - Kapoor et al, Journal of Pediatric Hematology Oncology, November 1995).

Treatment Refusal and Abandonment

The outlook of childhood cancer has improved enormously in the developed world with 5 year survival overall ~80%. Unfortunately we have not been unable to translate this success to much of the developing world including India. Treatment related mortality and relapse are important causes of treatment failure. However they are overshadowed by the widespread prevalence of treatment refusal and abandonment which are the main causes of treatment failure. These are as much a social problem as they are a medical problem. Any attempts to improve the outlook of children with cancer in a country like India has to have at its core strategies to deal with this problem. Such successful models are now being increasingly reported from Central and South America. Most notable has been the reduction of treatment abandonment in children with acute leukaemia as a result of the “twinning programs” between St. Jude Children’s Research Hospital in Memphis USA and Instituto Materno Infantil de Pernambuco in Recife, Brazil (see graph below)

Ref - Howard et al, JAMA, May 2004

Attempts have been made in the past to contact the children and families of those who abandon treatment. There are a very few studies looking at this and their dissemination has been limited to scientific proceedings. The recently published paper from Indonesia (Ref - Sitaresmi et al, Psycho-Oncology, April 2010) is the first extensive analysis of child and parent attitudes and behaviour to explain this complex problem. The authors conducted home-visits to interview families of 37 acute leukaemia patients, diagnosed between January 2004 and August 2007, who refused or abandoned treatment. There was no relation of abandonment with age, risk stratification of leukaemia, class of hospitalization (marker of economic status), parental education level and travel time. Girls abandoned slightly more often but this difference did not reach statistical significance.

Financial difficulties and a belief in the incurability of acute leukemia were the leading reasons given by parents for abandoning treatment. Additionally, treatment-related toxicity, painful procedures performed with inadequate analgesia and sedation, inadequate communication provided by health care providers and transportation difficulties were contributing reasons.