In addition to the known sex-specific cancer types, such as ovarian and prostate cancers, there are significant gender disparities in non-sex-specific cancers, such as rate of incidence and susceptibility, tumor aggressiveness, prognosis, and treatment response. A recent study published in Cancer Cell explored the molecular basis for these differences (Cancer Cell. 2016;29:711-722).
Although the medical research community has long understood that gender can influence the development and behavior of tumors in patients, this study provides a systematic, molecular-level analysis of gender differences as they present themselves in diverse cancers. It underscores the potential for developing sex-specific therapeutic strategies to prevent and treat certain cancer types, including those of the thyroid and head and neck.
Among the cancer types analyzed, the study’s findings identified two sex-effect groups of cancers associated with distinct incidence and mortality profiles: a strong sex-effect group that showed extensive sex-biased molecular signatures and a weak sex-effect group that contained a small number of sex-affected genes. More than 50% of clinically actionable genes showed sex-biased molecular signatures in certain cancer types.
Strong sex-effect cancers identified in the study included those of the thyroid and head and neck, as well as bladder, liver, and two types each of lung and kidney cancers. Cancers in the weak sex-effect group included specific brain tumors (glioma and glioblastoma), colon and rectum cancers, and acute myeloid leukemia.
Previous studies that have reported sex-related molecular patterns have been limited to individual genes, single molecular data types, and single cancer lineages. Another recent study identified H3K27me3 demethylase UTX as a gender-specific tumor suppressor in T-cell acute lymphoblastic leukemia, a cancer that has shown a skewed distribution toward males (Blood. 2015;125:13-21).
With this study, researchers sought to advance understanding of gender differences with a systematic, multidimensional analysis of sex-affected genes. “So, you take two patients with oropharyngeal cancer—although they didn’t look specifically at patients with oropharyngeal cancer—they hypothesize that for this and other cancers the outcomes may be very different for a male versus a female,” said Randal S. Weber, MD, FACS, professor in and chair of the department of head and neck surgery at the University of Texas MD Anderson Cancer Center (MDACC) in Houston. “What they found is that their tumors have different genetic signatures or profiles, so it may explain why a male’s prognosis might be different from a female’s.”
For the study, researchers used molecular data available through The Cancer Genome Atlas (TCGA) Project to perform a comprehensive, rigorous, pan-cancer analysis. The goal: to address whether or not there are molecular-level differences between the male and female cancer patients who have otherwise similar clinical and tumor characteristics and, if so, to identify them.
The team analyzed molecular-level differences between approximately 3,200 males and females with cancer using an analytic approach based on the propensity score, a statistical tool that allowed researchers to adjust for confounding factors such as age, race, disease stage, and tumor purity. Investigators focused on 13 major cancer types, including head and neck squamous cell carcinoma (HNSCC) and thyroid carcinoma, for at least five out of six molecular data types.
Examining the global patterns of sex-biased genes across different molecular types, the investigators found a clear separation among the cancer types under survey, with the cancer types in the strong sex-effect group showing a higher cancer incidence sex-bias index and a higher mortality sex-bias index. Their findings were reinforced by the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology, which suggest a patient’s sex as a prognostic factor in five of the eight strong sex-effect cancers named in the study.
Researchers focused on highly mutated genes in each cancer type and identified specific biases toward males or females. Study authors acknowledge, however, that their analysis based on the mRNA expression data of related TCGA normal samples “detected much fewer sex-biased genes, suggesting the sex bias might be amplified during the tumorigenesis process.” They also go on to point out that the so-called normal tissues analyzed consisted of distinct cell types from the corresponding tumor samples, thus potentially confounding the observable tumor-normal differences. In addition, the normal tissues’ sample size was significantly smaller than that of the tumor samples, possibly limiting “detection power.” The authors conclude that further efforts are required to shed more light on the “relative contributions of various factors (e.g., sex chromosomes, hormones, and tumorigenesis) to the observed sex-biased gene expression signatures in cancer samples.”
Thyroid, and Head and Neck Cancers
Thyroid cancer, one of the study’s identified strong sex-effect cancers, is already known to have a gender bias. A review by Yao and colleagues discusses results of the epidemiological, clinical, and experimental research on the role of sex hormones, their receptors, and other molecular factors in the higher incidence of thyroid cancer in women (Expert Rev Endocrinol Metab. 2011;6(2):215-243).
“We see that papillary thyroid cancer is four times more common in women than in men,” said Erich M. Sturgis, MD, MPH, professor in the departments of head and neck surgery and epidemiology at MDACC. “You would assume the underlying reasons for that might suggest that the mutation patterns in men and women are different because the background in which it arises is different and that attributing factors, such as hormone level, are an important component of thyroid cancer.”
Gender-related factors in head and neck cancers are not as broadly proven or recognized, and this study suggests significant differences at the molecular level between male and female patients of head-neck and thyroid cancers, said Han Liang, PhD, associate professor in the department of bioinformatics and computational biology at MDACC and one of the study’s authors.
Dr. Sturgis noted that specific forms of head and neck cancers were not separated out for analysis. “Head and neck cancer is not a singular disease,” he said. “The TCGA includes oral cavity cancer, oropharynx cancers, and laryngeal cancers, and those are really three different diseases.” Dr. Liang states that although the grouping did consist primarily of tumors from the oral cavity, oropharynx, and laryngeal site—samples from the TCGA—these cancer forms were not looked at separately “due to the limited power of smaller sample size.”
There are known gender incidence differences between certain head and neck cancers, Dr. Sturgis said, although the reasons for those differences are not clearly understood. “HPV-related head and neck cancers are about four to five times more common in men than women,” he said. “Now, that may just be behavioral differences between men and women and, thus, differences in exposure. But the idea that hormonal factors and other differences between men and women influence that is certainly possible.”
To put it in context, this study is part of the greater medical movement toward precision medicine. “Currently, male and female patients with many cancer types are often treated in a similar way without explicitly considering the factor of sex,” said Dr. Liang. “While this practice may be appropriate for the cancer types in the weak sex-effect group, special consideration should be given to those in the strong sex-effect group in terms of both drug development and clinical practice.”
Otolaryngologists already take gender into account in certain instances, noted Dr. Sturgis, but on a limited basis. “You may be a little more suspicious about the possible presence of HPV-related tonsil cancer in a white man in his 50s than a white woman in her 20s, because the latter is from an unusual group for getting HPV-related tonsil cancer,” he said. “So, certainly it influences us, but the signs and symptoms are what drive the diagnostic workup.”
It is in the treatment aspect that studies such as these will ultimately come into play—but not quite yet. Both Drs. Sturgis and Weber characterize the study’s findings regarding thyroid and head and neck cancers as “not ready for prime time” when it comes to making decisions about cancer treatment. “You would need to have the specific head and neck cancers separated out but, more importantly, we really need to start gathering genetic data, meaning genomic data, on tumors prospectively, so we can compare outcomes with different treatments between tumors that have different genomic patterns,” Dr. Sturgis added.
Delineating strong sex-effect cancers may give practitioners an additional tool and “is on the continuum of our efforts to develop personalized cancer therapy,” said Dr. Weber. “The study shows that there are genetic differences between some of the same cancers in males versus females, and you should account for this as you develop personalized therapy. Simply stated, the drug you give to a male for the same cancer as a female may not produce the same outcome.”
Understanding the genetic basis for cancer also gives practitioners the benefits of enhanced information and context when consulting medical literature on thyroid and head and neck cancers. It will bolster their ability to spot “targets of opportunity,” said Dr. Weber, an approach that is already being successfully used in many cancer treatments. “We have immunotherapy opportunities where we can reverse the immunosuppressive effects of tumors and increase the activity of the immune response to the presence of a tumor, and we have drugs that can target processes that can drive cancer progression,” he added.
This study is a stone in the pathway to providing an ultra-targeted clinical treatment for each type of cancer affecting each individual person. “One of the important things about going in this direction, especially regarding sex differences, is that these treatments are really, really expensive,” said Dr. Weber. “And, until we have a method or assay to inform us as to who might respond and who might not, we give these expensive drugs to 100 patients in hopes that it might work for a third of them. With predictive molecular analysis, and analysis of response to a treatment, we can give patients the right drug at the right time for the right reason. This is really what precision medicine is about.”