Immunoscore

How it is measured. Paraffin sections of the tumour and its invasive margin are stained for CD3 and CD8, the densities are quantified by digital pathology, and the score is the mean of four density percentiles, reported as low, intermediate or high.

The validation. An international consortium of 14 centres in 13 countries, led by the Society for Immunotherapy of Cancer, assessed the assay in stage I to III colon cancer. Samples from 3,539 patients were processed and 2,681 passed quality control, split into training (700), internal validation (636) and external validation (1,345) sets. Reproducibility between observers and centres was high (correlation 0.97). In the training set, recurrence at five years occurred in 8 percent of patients with a high score, 19 percent with an intermediate score and 32 percent with a low score (hazard ratio 0.20 for high against low), and the finding held in both validation sets. The association with time to recurrence was independent of age, sex, T stage, N stage, microsatellite instability and the existing prognostic factors, and in the 1,434 patients with stage II disease the difference remained significant (hazard ratio 0.33). The Immunoscore made the largest relative contribution to recurrence risk of any clinical parameter, including the stage itself (Pages 2018).

What it is for. The clinical question it speaks to is which stage II patient needs adjuvant chemotherapy, where stage alone decides badly. It has not displaced stage or circulating tumour DNA in routine practice, and it is measured in research and in some European centres rather than reported as standard. The immune biology it measures is the same one that makes mismatch repair-deficient tumours respond to checkpoint inhibitors.