Most tumours of non-reproductive tissues are known to exhibit biological sex differences, with both incidence and mortality rates being higher in males. Sex disparities also extend to other aspects of cancer biology and treatment. Therefore, understanding the molecular mechanisms of sexual dimorphism in cancer is paramount to improving patient care. Two studies have now provided evidence of differences, dependent on patient sex, in oncogenic features, such as frequencies of driver gene mutations, mutation load and mutational signatures, as well as immune selection.
Li et al. used data from the PCAWG Consortium to look for sex differences in mutations from 1,983 samples of 28 tumour types, omitting those that derive from reproductive tissues. By first focusing on mutation frequencies of particular driver genes, the authors were able to confirm two previous findings: mutations in CTNNB1, encoding β-catenin, were more common in hepatocellular carcinomas (HCCs) originating in males, and TERT promoter non-coding mutations were more prevalent in thyroid cancers of males. Furthermore, assessing mutation burden across all the pan-cancer samples revealed that the overall number of somatic single nucleotide variants (SNVs) was greater in tumours of males than females. Specifically, male-derived HCCs and renal cell carcinomas were associated with a higher prevalence of SNVs, while male-derived thyroid cancers trended towards a higher number of coding mutations.
Other interesting features of tumour evolution were identified in biliary adenocarcinoma and were associated with a sex bias. These included polyclonality — tumours originating from females were more often polyclonal compared with monoclonal male-derived tumours — and the mutation timing of structural variants (SVs) — with the proportion of truncal SVs more common in male-derived tumours than female ones.
To examine whether underlying mutational processes might account for the sex disparities observed in mutation burden and tumour evolution, comparisons of mutational signatures between the sexes were made. Across all of the tumour types, three mutational signatures were detected that were more common in one sex over the other. SBS1, a base substitution signature arising from deamination of 5-methylcytosine to thymine, was more frequently observed in tumours of females whereas two further base substitutions, of yet undetermined aetiology, SBS17a and SBS17b occur more often in tumours of males. Further work will be needed to establish whether lifestyle or biological factors accounting for these mutational signatures can explain the molecular sex differences.
In another study, Castro et al. sought to explore why, in clinical trials of multiple cancer types, responses to immune checkpoint blockade (ICB) are inferior in young and female patients compared to older and male patients. Seemingly this observation contradicts previous reports demonstrating that females display stronger immune responses than males in other contexts including having increased CD4+ T cell numbers. The hypothesis put forward by the authors was that young and female patients retain fewer mutations that can be efficiently presented by their own MHC molecules. This might result from stronger immunoediting early in tumour evolution wherein the immune system eliminates tumour cells presenting more immunogenic mutant peptides.
To test their hypothesis, the authors’ used a method previously developed by their group called the Patient Harmonic-mean Best Rank (PHBR) score, which is a measure of the capacity of a patient’s specific MHC I and MHC II haplotypes to present peptides from their tumours harbouring somatic mutations. The PHBR score was calculated for 1,018 driver mutations using >7,100 patients with microsatellite-stable tumours from The Cancer Genome Atlas (TCGA) dataset. The patient cohort was diverse in terms of sex, age and tumour type and importantly, patients with tumours of reproductive tissues were excluded from analyses of sex differences.
After determining PHBR scores across all the tumour types, females were found to have higher PHBR scores than males, which indicates a decreased ability to present driver mutations through their MHC I and MHC II molecules and higher selective pressure. This worse presentation of driver events was also observed in younger patients compared to older patients. Furthermore, tumours from younger females had the strongest immune selection relative to the other patient cohorts — older females, younger males and older males — suggesting that sex and age have additive effects. Importantly, these findings could be confirmed using an independent non-TCGA patient cohort. Overall, this study proposes that females and younger individuals are less able to mount an efficient host antitumour response as depletion of immunogenic mutant peptides makes their tumours less visible to the immune system, potentially offering an explanation for the ineffectiveness of ICB in this patient population.
“differences, dependent on patient sex, in oncogenic features”
With standard cancer care yet to factor in patient sex, these two studies call for increased recognition of sex differences in cancer research and clinical management.
Castro, A. et al. Strength of immune selection in tumors varies with sex and age. Nat. Commun. 11, 4128 (2020)
Li, C. H. et al. Sex differences in oncogenic mutational processes. Nat. Commun. 11, 4330 (2020)