On the Prognostic Power of Tumor-Infiltrating Lymphocytes - A Critical Commentary

Tumor-infiltrating lymphocytes are extensively used as prognostic biomarkers in cancer. Regulatory T cells (Tregs) or CD8+ T cells frequencies in tumor site, or their ratio, are the most common markers used to assess prognosis. This work offers a possible explanation for the opposite correlations between intra-tumoral Tregs and survival, associated with different types of cancer. The complexity involved with the selection of a preferred marker, including the effect of variability, is presented and discussed. The lymphocytes frequency ratio is proposed as the marker of choice in most types of cancer. The ratio correlates directly with survival, irrespective of cancer type and is also less variable than the frequencies of each of the two lymphocytes, if these frequencies correlate with each other in the tumor microenvironment. However, if the frequency of one of the two lymphocytes is highly variable, abandoning the ratio in favor of the lymphocyte with less variable frequency will improve correlation with survival, especially when the intra-tumoral frequencies of the two species are inversely correlated. It is plausible, that the best prognostic marker selected this way, will be also be the best predictor of checkpoint inhibitor therapy success.

Automated summary

Tumor-infiltrating lymphocytes, particularly the frequencies of regulatory T cells (Tregs) and CD8+ T cells, are important prognostic markers in cancer. The correlation between intra-tumoral Tregs and survival can vary depending on the type of cancer. This study suggests using the frequency ratio of lymphocytes as the preferred marker in most cancer types, as it shows a direct correlation with survival and is less variable compared to individual lymphocyte frequencies. However, if the frequency of one lymphocyte is highly variable, it may be beneficial to focus on the lymphocyte with a less variable frequency to improve the correlation with survival, especially when the intra-tumoral frequencies of the two species are inversely correlated. The marker selected using this approach may also be a good predictor of checkpoint inhibitor therapy success.