Knowing how far a cancer has spread at diagnosis is essential for planning treatment, estimating outcomes, and tracking cancer trends. However, many cancer registries record incomplete stage information, and different registries use different staging systems, making it hard to compare data across hospitals, regions, and countries.
At the Indian Council of Medical Research–National Institute of NCD Epidemiology (ICMR–NINE), Bengaluru, we built a framework to translate between four widely used staging systems: the detailed TNM system, and three simpler systems—SEER Summary Stage, Essential TNM, and Condensed TNM. We harmonized these systems for breast, cervix, oral cavity, and colorectal cancers by comparing the official criteria of each system category by category against TNM, mapping which categories in one system correspond to which in another. We then tested this map using real patient records from 242 Hospital-Based Cancer Registries across India (2018–2023), covering 720,539 patients, of whom 315,470 had complete staging details. A representative sample of 4,000 patients let us check how accurately stage could be converted between systems, in both directions.
We found the translation worked well for some cancers but not others. Cervical and colorectal cancers showed strong agreement across all systems, meaning registries could reliably substitute one system for another. Breast and oral cancers showed more disagreement, particularly at intermediate stages, because the simpler systems do not always weigh tumour size and lymph node spread the way TNM does.
Through this work, we have given Indian and other low- and middle-income country registries a validated, cancer-specific tool to convert stage data between systems rather than discarding incomplete or differently coded records. We believe this can strengthen public health surveillance by enabling more complete monitoring of stage at diagnosis, better tracking of trends over time, and pooling of data across registries and countries for research and international comparisons. For clinical care, our framework does not replace individual patient assessment, but by improving the completeness and comparability of population-level staging data, it can strengthen the evidence used to plan cancer-control programmes, allocate screening and treatment resources, and identify gaps in early detection. Our findings also show that reliable conversion needs detailed tumour, node, and metastasis information, not just the final stage label.