#AACR16 Update: How Immune Infiltration Predicts Survival in Cancer Patients

The Immunoscore—a method of characterizing the nature and function of immune cell infiltrates into tumors—is the best predictor of survival in colorectal cancer (CRC) patients, according to Jérôme Galon, PhD, who in a late-day talk yesterday described how the Immunoscore’s classification system can inform therapeutic strategies for patients.

Galon, who received CRI’s 2010 William B. Coley Award for pioneering this unique prognostic approach, recently showed in a group of over 800 patients that higher Immunoscores–which reflect the presence of a pre-existing immune response in and near tumors– are associated with better survival in patients, regardless of tumor stage or whether they had invaded surrounding tissues. This predictive power also holds true for brain metastases, where patients with high Immunoscores survived at twice the rate of patients with low Immunoscores.

These insights offer clinicians a new perspective on tumor progression, and have especially important implications for treating CRC with immunotherapy.

Checkpoint inhibition immunotherapy has led to remarkable responses in patients with heavily mutated tumors, especially melanoma and lung cancer, because mutations can make cancer cells look foreign to the immune system. Therefore, the more mutations a tumor has, the more potential targets the immune system can zero in on.

This rationale led to a recent phase II study in which the anti-PD-1 antibody pembrolizumab proved highly effective in a subset of patients with advanced CRC, whose tumors had high levels of microsatellite instability (MSI). The presence of MSI indicates a malfunction in a cell’s DNA repair system, allowing mutations to accrue over time. Of the ten patients with MSI-high tumors, nine either responded or experienced disease control. In contrast, immunotherapy benefited only two of the eighteen patients with non-MSI-high tumors.

The MSI-high tumors were characterized by expression of PD-L1 and infiltration of CD8+ T cells, and researchers concluded that MSI status was the primary determinant of patient survival, as well as indicative of which tumors were likely to respond to checkpoint inhibition. However, upon application of the Immunoscore to over 800 CRC cases, Dr. Galon showed that while MSI was a good predictor of survival, the Immunoscore proved a superior prognostic method.

The high level of mutation in MSI tumors can cause immune cells to infiltrate and attack cells within the tumor, and thus those patients tend to have better survival. But patients whose MSI tumors had low Immunoscores, and hence low immune cell infiltration, fared no better than patients with non-MSI tumors with similarly low Immunoscores. On the other hand, non-MSI tumors with high Immunoscores were associated with the same improved survival as patients with immune-infiltrated MSI tumors. Therefore, MSI status correlates with better prognosis only to the extent that it is also accompanied by immune cell infiltration.

The Worldwide Immunoscore Consortium, which Dr. Galon directs, is seeking to further validate the Immunoscore metric through a massive effort consisting of 23 international centers that are currently applying the Immunoscore criteria to over 3,000 CRC tumors.

To end his talk, Dr. Galon spoke of how the Immunoscore should guide treatment approaches for CRC. For patients with high Immunoscores, checkpoint inhibition is the logical strategy, because there are already immune cells in the tumor that just need to be unleashed. On the contrary, for patients with low immunoscores, effective treatment approaches will have to focus on getting immune cells into the tumor in the first place, before they can exert their anti-tumor activity and improve patient outcomes.