Discussion
Data on the spectrum of CNS tumors remains to be consistent with studies in India and in the West. Medulloblastoma, low- grade glioma and ependymoma are the most common CNS tumors. However, very few optic pathway tumors were diagnosed, as compared to the UK, where optic pathway tumors make up 5.2% of all tumors. A possible explanation for this could be that these tumors are diagnosed and treated elsewhere, for example by ophthalmologists, reflecting the fragmented nature of care in India. Furthermore, this study detected a greater proportion of high grade/ metastatic tumors compared to low grade tumors. Tumors that did not need any form of treatment or chemotherapy were not reported in this study.
Low-grade tumors have significantly higher TDI than high grade tumors. This is consistent with past research, which concluded that aggressive tumors had relatively shorter TDIs 29. This is due to the rapid onset of symptoms, leading to a quicker diagnosis. On the other hand, slow growing tumors take longer to be diagnosed and hence are identified at an advanced stage. Low-grade gliomas were seen to have one of the highest TDIs in the UK and South India, where the SI was twice as long as the PI. This could possibly be due to lack of proper referral pathways or the failure to detect signs and symptoms as being related to a brain tumour by the initial HCPs.
Tumours located on the midline had the lowest TDI of 2.1 weeks, which is strikingly different to the UK data which concluded that anatomically midline tumors were associated with longer TDIs (Table 2). Inversely, cerebellar tumors had one of the longest TDI of a median of 4.1 weeks, whereas cerebellar tumors were seen to have one of the shortest TDIs of 7.4 weeks. Whilst we cannot ascertain the reason behind this discrepancy, we will review this in future surveys. Table 2 shows the differences between TDIs and PIs the HeadSmart data in the UK and this study findings based on tumour type.
All patients received a Magnetic Resonance Imaging (MRI) scan the same day they arrived at the hospital, which indicates good imaging infrastructure in the specialist centres. The age-specific differences in TDI are interesting. Children <5 years of age were noted to have relatively short TDI, but longer PI. This could be explained by the inability of young children to clearly express how they are feeling and can often lead to a substantial delay. Older children aged 12 and above were seen to have the highest TDI, with a significantly longer SI. Adolescent patients are present distinct psychological and physiological challenges that can lead to certain signs and symptoms going undetected and this is in keeping with previous studies
Patients who lived in the city had a significantly shorter TDI, PI and SI as compared to patients who lived in villages. However, the correlation between distance from home and hospital where the first diagnosis was made is a bit unclear. The risk of visiting more HCPs prior to diagnosis increases with distance as patients living more than 20km away from the hospital are 8.6 times more likely to have visited more HCP prior to diagnosis.
Signs and symptoms are consistent with UK data. Special consideration must be given to recognizing motor and visual signs. Earlier studies indicate the difficulty of assessing these signs in children. Furthermore, behavioural signs such as lethargy are also hard to assess, but most also be given importance as they are the most common behavioural sign of tumour 12. A further study on the most common signs and symptoms presented by specific tumors in India could help raise awareness to HCPs and could aid in reducing SI for some tumors.
Since most of the data was collected from retrospective notes from the doctors, there is likely to be some recall bias. Since most hospitals in this experiment were privately owned or run by Non- Governmental Organizations (NGOs), it does not account for cases diagnosed at public institutions. There were also social factors where the patients may not have wanted their doctor to know that they visited more doctors thereby under-reporting HCPs seen.
A prospective study or data collection from cancer registry would be the ideal way obtaining this type of data. However, there is no national registry data that contains this information and this survey provides key baseline data upon which further awareness programmes to promote early diagnosis can be evaluated.
The strategies needed to overcome identified barriers should involve proper education and training of healthcare workers and involve establishing clear referral pathways. It is crucial for all paediatricians and primary healthcare providers to be sensitized to signs of cancer in children. There have been initiatives in India to aid this. A National Training Project under the Indian Association of Paediatrics IAP PHO Chapter has played an important role in promoting early detection and referral for paediatric cancer cases30. HCPs not only should be trained but should also have guidelines that can aid them in diagnosing brain cancers earlier. HeadSmart is a campaign established in the UK, which assists HCPs in the assessment of children who may have brain tumours14. This guideline has drastically helped reduce diagnostic intervals of paediatric brain cancers in the UK. A similar guideline tailored to the Indian system can greatly benefit HCPs in India.
Another strategy to combat delay in diagnosis is to establish a clear referral network through government health care policies. Current protocols such as MCP-841 has improved overall survival of Acute Lymphocytic Leukaemia (ALL) 31. Establishment of a proper referral pathway for paediatric brain cancers can significantly benefit patients by reducing patient and diagnostic intervals, which can lead to quicker treatment.
Additionally, more resources should be invested to incorporate holistic care for all patients. A study in TATA Memorial Hospital concluded that holistic patient support, which involved Accommodation, Nutritional and Educational support, made a measurable impact on children with haematological malignancies32. This can aid in reducing the financial burden and help prevent treatment abandonment.
Overall, the diagnostic intervals for paediatric brain tumors were comparable to data in the UK. Moreover, all patients received an MRI within a day, indicating that infrastructure was not a barrier in this study. However, many low-grade and optic pathway tumour were unaccounted for. A possible explanation could be that optic pathway tumour could have been treated at an ophthalmologist setting. Tamil Nadu has one of the best healthcare systems in India, meaning that the data presented cannot be extrapolated to other states or to the whole country. Extending this methodology to other areas with poorer healthcare provisions, could provide a better understanding of diagnostic intervals at a national level.