Molecular pathology selected abstracts

Create PDF

Editors: Donna E. Hansel, MD, PhD, chair of pathology, Oregon Health and Science University, Portland; Richard D. Press, MD, PhD, professor and director of molecular pathology, OHSU; James Solomon, MD, PhD, assistant professor, Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York; Sounak Gupta, MBBS, PhD, senior associate consultant, Mayo Clinic, Rochester, Minn.; Fei Yang, MD, assistant professor, Department of Pathology, OHSU; Andrés G. Madrigal, MD, PhD, molecular genetic pathology fellow, Department of Pathology, OHSU; and Erica Reinig, MD, assistant professor and medical director of molecular diagnostics, University of Wisconsin-Madison.

Affect of treatment on association between TMB and prognosis

March 2021—High tumor mutational burden in certain cancers has become an established biomarker for predicting a response to immune checkpoint inhibitor therapy and longer overall survival after such treatment. The immune checkpoint inhibitor (ICI) pembrolizumab, for example, has recently been approved by the FDA for patients whose solid tumors, regardless of histology, have a high tumor mutational burden (TMB), defined as 10 or more mutations per megabase. TMB, assessed by next-generation sequencing, varies considerably among cancers and can range from 0.01 to more than 1,000 somatic mutations per megabase of sequenced genome. The presumed mechanism for the enhanced responsiveness to immunotherapy associated with high TMB is the creation, by somatic mutation, of potentially immunogenic neoantigens that facilitate an enhanced antitumor immune response. Given this presumed mechanism, the authors addressed whether high TMB levels, which are associated with better cancer outcomes in patients treated with immune checkpoint inhibitors, might also lead to better outcomes for patients treated with other anti-cancer therapies. More specifically, is TMB a good generic prognostic biomarker for all cancer patients, regardless of therapy, perhaps due to a neoantigen-driven enhanced anti-cancer immune response? Or is it only a good prognostic marker in patients treated with immune checkpoint inhibitors that have been designed to enhance the anti-tumor immune response? The authors, all from Memorial Sloan Kettering Cancer Center, tackled these questions using an institutional retrospective cohort of 10,233 patients who had 17 different tumor types. All of the tumors were profiled by Memorial Sloan Kettering’s NGS-based MSK-IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets) assay. Eighty percent of these patients had not received ICI therapy. The authors used this genomic data, clinical and treatment information from patients’ medical records, and a sophisticated statistical model that considered not only the use of ICI therapy but the timing of that therapy in their study. With this information, they assessed the effects of TMB status on overall survival in patients with ICI treatment as opposed to other cancer treatments. They found that, in the absence of ICIs, high TMB status is a biomarker that imparts an overall poor cancer prognosis. The prognosis for TMB-high patients improves after treatment with ICIs. Among the 8,356 patients with microsatellite-stable tumors who had not received ICIs, high TMB was associated with worse overall survival, with a 26 percent greater risk of death compared to those with low TMB. Among all patients who received ICIs, high TMB was associated with improved overall survival, with a 26 percent decreased risk of death compared to those with low TMB. Because the FDA-defined 10 mutations per megabase TMB cutoff to indicate pembrolizumab responsiveness in any solid tumor has been controversial, the authors defined high TMB as the top 20th percentile within a cancer type. The association of high TMB status with worse cancer prognosis in non-ICI–treated patients was not observed in all cancer types. In colorectal cancer, endometrial cancer, and bladder cancer, but not other cancer types, high TMB was associated with better prognosis regardless of whether patients received ICIs. These three cancer types are known to often have defects in DNA mismatch repair, as measured by high levels of microsatellite instability, a biomarker that imparts a relatively good prognosis. Because microsatellite instability is a major contributor to overall TMB in these three tumors, the association of high-TMB status with a relatively good prognosis in these tumors may be driven by microsatellite instability-related mechanisms. These exceptions notwithstanding, the authors concluded that, in the absence of ICI therapy, TMB is a generic biomarker for poor prognosis in most cancers. However, the resulting mutation-induced increase in neoantigen formation represents an opportunistic target for intervention with ICIs to significantly improve overall survival.

Valero C, Lee M, Hoen D, et al. The association between tumor mutational burden and prognosis is dependent on treatment context. Nat Genet. 2021;53:11–15. https://doi.org/10.1038/s41588-020-00752-4

Correspondence: Dr. Venkatraman E. Seshan at seshanv@mskcc.org

Comparison of the genomes of identical twins

The identical phenotypic appearance of monozygotic twins and the twins’ derivation from the same fertilized egg have long led scientists to assume that their genomes are identical. Differences in their phenotypic characteristics, such as in disease development and behavior, have often been attributed to differing environmental exposures instead of to genetic or hereditary factors. The authors conducted a study of genomic differences in monozygotic twins in which they assessed the genomes of 381 pairs of such twins. The study showed that twin pairs, on average, have 5.2 mutational differences at birth, each of which occurred early in fetal development. In many of these twin pairs, some mutations are carried by nearly all cells in one twin but absent in the other. The study also showed that in about 15 percent of identical twin pairs, one twin carries a substantial number of mutations—10 to 15—that the other twin does not share. To assess the origin of these mutational differences between twins, the authors performed whole genome sequencing on 381 pairs of identical twins and their children, spouses, and parents. Sequencing of non-twin family members was necessary to determine whether a mutation found in only one twin was present in that person’s germline cells, rendering it transmissible to offspring, and/or his or her somatic cells. Unique mutations present in the germline and somatic cells of a twin most likely occur before primordial germ cell specification, an early developmental event that typically happens one to two weeks after blastocysts are formed. Twinning occurs when a single fertilized egg, or zygote, splits and develops into two embryos. This typically occurs between one and seven days after fertilization but can occur up to day 13. The later the split, the more cells will have accumulated and the higher the probability for mutations to have occurred in some of those embryonic cells prior to separating. The authors found 23,653 mutations, in total, that were specific to one twin, with a median of 14 postzygotic mutations that differed between twin pairs. However, there was considerable heterogeneity in the twin cohort, with 39 twin pairs having more than 100 mutations that were not shared. By studying family members to track the cellular origin of the mutations, the authors also unexpectedly found that in many twin pairs, some mutations are carried by nearly all cells in one twin but absent in the other. This suggests that one of the twins was formed solely from the descendants of the early embryonic cell where the mutation took place and the other was formed solely from the descendants of other nonmutated cells of the developing embryo. In some instances, a unique mutation was found in 20 percent of the cells in one twin and 100 percent of the cells in the other. In these cases, one twin was formed solely from the descendants of the early embryonic cell where the mutation took place and the other was formed in part by descendants of that same mutant cell and in part by other embryonic cells. It is likely that most of these twin-specific mutations are not disease associated, but some may be. Therefore, additional studies are needed to better assess the detailed pathogenicity of these twin-specific mutations. If some of these mutations are pathogenic for specific disease phenotypes, it may be necessary to reassess past twin studies and reconsider how to analyze future studies knowing that a disease outcome specific to one twin, but not the other, may not result from different environmental exposures.