It would be hard to identify an approach to cancer treatment that has received more attention recently than anti-checkpoint therapy (Pollack, 2015).  This strategy for eliminating tumor cells is based on interfering with one or another pathway that inhibits the initial activation or functions of T cells, such as CD8+ cytotoxic T cells (CTL).  Activated tumor-specific CTL can directly kill their targets.  However, if copies of the T-cell surface molecule, PD-1, are bound by their physiological ligands on tumor cells, either PD-L1 or PD-L2, or other cells the ability of the T cell to perform its functions is substantially reduced.  A report published in Science (2015) by Rizvi et al. last month addresses the question of whether tumor mutation burden correlates with response to anti-checkpoint therapy for non-small cell lung cancer (NSCLC).

The authors studied two cohorts of NSCLC tumors, 16 samples in a discovery cohort and 18 in a validation cohort.  They subjected these samples to DNA sequencing with comparable coverage and depth for relevant genes in the two cohorts, discovery and validation, as well as for patients who did or did not exhibit clinical benefit.  A clear and strong correlation was seen for patients in either cohort between the numbers of nonsynonymous (NS) tumor mutations and the extent of response to therapy with the human monoclonal antibody, pembrolizumab, which binds to PD-1 and prevents the interaction between PD-1 and PD-L1 or PD-L2.

For example, for discovery cohort tumors from patients with a durable clinical response, the average number of NS mutations was 302 versus an average of 148 NS mutations for tumors from patients without a durable benefit.  Patients with a high tumor NS mutation burden also had substantially better objective response rates and progression-free survival than patients with low tumor NS mutation burden.  These correlations were considerably weaker for total exonic mutation burden (which would include synonymous mutations that would not generate new tumor-specific antigens), a finding consistent with an immunological mechanism for the impressive relationship between tumor NS mutation burden and clinical benefit.

A previous study demonstrated that smoking is associated with a preponderance of C-to-A transversion mutations as opposed to C-to-T transition mutations in tumors.  For the tumors investigated in the present study, better durable clinical benefit, objective response rate, and progression-free survival were found in the tumors with the smoking-associated mutation pattern versus the pattern not associated with smoking.  This result also supports the notion that NSCLC tumors with greater numbers of NS mutations are more likely to be effectively treated by pembrolizumab, as smoking has been shown to be associated with increased numbers of mutations in tumor cells.

The authors then addressed whether the greater number of NS mutations in tumors from patients with better clinical responses to pembrolizumab was associated with greater numbers of neoantigens that could serve as targets for T-cell mediated immunity.  Consistent with the hypothesis motivating this analysis, tumors from patients with durable clinical benefit exhibited a greater number of neoantigens than tumors from patients without durable clinical benefit.  Rizvi et al. also demonstrated that peripheral blood CD8+ T cells from a NSCLC tumor patient could be found to exhibit cytotoxicity against target cells displaying mutant but not wild-type peptides in the context of self-MHC class I molecules.  The kinetics of this T-cell−mediated reactivity correlated with the kinetics of tumor regression.

Previously, Snyder et al. (2014) obtained similar results to those of Rizvi et al. by studying melanoma patients with and without clinical benefit following treatment with monoclonal antibodies to CTLA-4, another T-cell surface molecule and target of anti-checkpoint therapy.  In these melanoma patients, mutational load was correlated with the degree of clinical response to therapy.  The mechanism underlying this relationship once again appeared to involve responses by T lymphocytes to tumor neoantigens.  Work by Llosa et al. (2015) is also consistent with a correlation between the NS mutation load of colon cancers and clinical benefit from anti-checkpoint therapy, with better clinical responses for colorectal tumors characterized by microsatellite instability, which generally exhibit increased numbers of mutations in comparison to colorectal tumors without microsatellite instability.

References

Pollack A. New class of drugs shows more promise in treating cancer. New York Times, May 29, 2015.http://www.nytimes.com/2015/05/30/business/new-class-of-drugs-shows-more-promise-in-treating-cancer.html?hp&action=click&pgtype=Homepage&module=second-column-region&region=top-news&WT.nav=top-news&_r=0.

Rizvi NA, Hellmann MD, Snyder A, Kvistborg P, Makarov V, Havel JJ, Lee W, Yuan  J, Wong P, Ho TS, Miller ML, Rekhtman N, Moreira AL, Ibrahim F, Bruggeman C,  Gasmi B, Zappasodi R, Maeda Y, Sander C, Garon EB, Merghoub T, Wolchok JD,  Schumacher TN, Chan TA. Cancer immunology. Mutational landscape determines  sensitivity to PD-1 blockade in non-small cell lung cancer. Science. 2015 Apr 3;348(6230):124-8. doi: 10.1126/science.aaa1348. Epub 2015 Mar 12. PubMed PMID: 25765070.

Snyder A, Makarov V, Merghoub T, Yuan J, Zaretsky JM, Desrichard A, Walsh LA, Postow MA, Wong P, Ho TS, Hollmann TJ, Bruggeman C, Kannan K, Li Y, Elipenahli C, Liu C, Harbison CT, Wang L, Ribas A, Wolchok JD, Chan TA. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014 Dec 4;371(23):2189-99. doi: 10.1056/NEJMoa1406498. Epub 2014 Nov 19. PubMed PMID: 25409260; PubMed Central PMCID: PMC4315319.

Llosa NJ, Cruise M, Tam A, Wicks EC, Hechenbleikner EM, Taube JM, Blosser RL, Fan H, Wang H, Luber BS, Zhang M, Papadopoulos N, Kinzler KW, Vogelstein B, Sears CL, Anders RA, Pardoll DM, Housseau F. The vigorous immune microenvironment of microsatellite instable colon cancer is balanced by multiple counter-inhibitory checkpoints. Cancer Discov. 2015 Jan;5(1):43-51. doi: 10.1158/2159-8290.CD-14-0863. Epub 2014 Oct 30. PubMed PMID: 25358689; PubMed Central PMCID: PMC4293246.