For some time, genetic testing has been predictive and prognostic. It is now assuming a therapeutic role as well. An example is the targeting of breast cancer patients with BRCA mutations for treatment with PARP inhibitors.
Hereditary cancer risk assessment has grown from a limited concentration on a few well-described tumors and genetic mutations to an ever-widening understanding of cancer syndromes encompassing multiple tumor types across multiple generations. The identification of an increasing number of genetic abnormalities and the advances in our understanding of the role of genetics in the risk of developing a cancer have created new opportunities to impact patient care. Primary prevention (the possibility of changing outcomes associated with risk) and secondary prevention (the ability to identify cancers at earlier stages, with the goal of intervening at a time at which outcomes can be changed in a meaningful way) are becoming more of a reality as it becomes possible to recognize hereditary cancer syndromes.
With the advent of more accessible testing come challenges of maintaining rigor when defining and categorizing results and determining their significance. The broadening of our definitions of risk has already led to inclusion criteria that did not previously exist, such as the association of BRCA mutations with an increased risk of ductal carcinoma in situ and triple-negative breast cancer. Further refinements in our understanding of the relationship between specific disease biology and germline genetics will result in new associations between types of cancer and risk.
Newly identified genetic mutations associated with hereditary cancer syndromes have recently been classified into levels of risk. Specific gene mutations are associated with high or moderate risk, and society guidelines have acknowledged these risk levels and made varying recommendations for screening and prevention. At the same time, however, we have come to recognize that not all similarly identified genetic mutations have the previously ascribed clinical risk. General wisdom, supported by the literature, was that TP53 mutations were highly penetrant, with the associated Li-Fraumeni syndrome resulting in diffuse and severe clinical cancer throughout the affected family-although it was acknowledged that this diagnosis, while serious, was quite rare. The widespread use of multigene panel testing has raised doubts about the preceding description being the single pattern of the gene’s expression. More cases of TP53 mutations are being identified, and the familial patterns observed often differ from the traditional one. Also as a result of more widespread use of multigene panel testing, we are beginning to elucidate specific genetic syndromes in which we can predict phenotype on the basis of specific genotype, such as with the varying mutations in mismatch repair genes seen in Lynch syndrome. The collection of more data in a controlled, organized manner will allow us to continue to perfect our understanding of the effects of particular genetic mutations on specific organs. There are already several areas in which we have begun to see this: 1) in crossover syndromes, where genetic mutations originally regarded as causing cancer in one set of organs now, with the collection of more data via multigene panel testing, have been shown to increase cancer risk in organs not previously thought to be involved (eg, the apparent increase in breast cancer risk in patients with mutated PMS2); and 2) in associations of new cancers with previously established gene patterns (eg, the presence of a high risk of uterine cancer in some BRCA1 patients, as described recently by Shu).
To determine how best to approach population risk analysis, we must look critically at our current risk models. Models such as Tyrer-Cuzick, BRCAPRO, Gail, and Claus take different factors into consideration and place differing emphases on those factors identified in multiple models. Typically, women above a defined threshold are deemed high-risk; however, using a continuum is a more appropriate approach to risk analysis, given that the models target specific populations and have not been validated in all populations or for all individuals.
As practicing clinicians, we often encounter situations where a patient is dropped into one or another “risk bucket” by the risk model calculation, yet our clinical judgment leads us to differ with this determination for that individual. Additionally, outcome data are often lacking for the generally accepted screening recommendations.
The development of a consensus in the medical community with regard to risk evaluation and medical management requires solid clinical data and will take time. New associations between genes and cancer risk will be identified, new mutations will be discovered, the significance of mutations already identified will be updated, and cofactors will be elucidated. From this bank of information, gene expressions can be tracked, interventions can be explored and validated, and ultimately, outcome studies of overall survival can be performed. Until these things are accomplished, we must rely on practiced clinical judgment.
Genetic testing is becoming commonplace. Practitioners from many medical specialties are ordering tests and making recommendations to patients. Testing in the United States is regularly ordered by gynecologists, primary care and family practice physicians, medical oncologists, surgical oncologists, genetic counselors, mid-level providers, and some radiology facilities. Each of these specialties serves patients and populations that may benefit from risk analysis and genetic testing. The challenge is to be sure that those offered testing are appropriate candidates, are counseled knowledgably, and, importantly, are afforded clinical follow-up once their risk level is determined. We have seen the heartbreak of situations where well-intentioned clinicians have offered testing in a medical setting without adequate follow-up in place.
Testing by medical providers is good when appropriate, but those ordering the testing need to understand the implications and have the ability to execute a plan of care. Genetic counselors are helpful in certain situations when available, but genetic counselors are not clinicians and do not have the ability to provide patient care-and sending the patient back to the referring physician is only adequate if that physician is versed in care of the particular clinical syndrome involved and knowledgeable about the associated risk. A clear and appropriate plan must be in place for the patient’s clinical follow-up. The challenge is to make appropriate risk assessment and genetic testing widely available, and also to give proper attention to the results. The ideal would be a coordinated team approach that includes clinicians, genetic counselors, and the patient’s medical providers, but this arrangement is a luxury found in only a few places in our country.
For some time, genetic testing has been predictive and prognostic. It is now assuming a therapeutic role as well. An example is the targeting of breast cancer patients with BRCA mutations for treatment with poly (ADP-ribose) polymerase (PARP) inhibitors. This is the tip of the iceberg. Targeted cancer therapy, determined by an understanding of the genetics of the particular tumors involved, is on the horizon. Moreover, it is not too far in the future that whole-genome testing will be a reality. This will have enormous therapeutic potential and will open new avenues of thought, based on a better understanding of ourselves and disease processes. Use of the information gleaned from whole-genome testing will depend on the integrity of the scientific and medical communities, and preparations need to be put in place to handle this information in the best way possible for each individual patient.
Financial Disclosure:Dr. Smith has served on the Speakers Bureau for Myriad Genetics. Ms. Sin has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.
1. National Comprehensive Cancer Network. Genetic/familial high-risk assessment: breast and ovarian. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). Version 2.2016. http://www.nccn.org/professionals/physician_gls/pdf/genetics_screening.pdf. Accessed September 16, 2016.
2. ten Broeke SW, Brohet RM, Tops CM, et al. Lynch syndrome caused by germline PMS2 mutations: delineating the cancer risk. J Clin Oncol. 2014;33:319-25.
3. Shu CA, Pike MC, Jotwani AR, et al. Uterine cancer after risk-reducing salpingo-oophorectomy without hysterectomy in women with BRCA mutations. JAMA Oncol. 2016 Jun 30. [Epub ahead of print]