This article is more than 5 years old. Information may no longer be current.
How precise is precision medicine, anyway?
Click Here to Manage Email Alerts
“By concentrating on precision, one arrives at technique, but by concentrating on technique one does not arrive at precision.”
— Bruno Walter
Precision medicine is arguably a sterile-sounding alias for personalized medicine.
One way or the other, both terms encompass the breadth of all objective and highly accurate means by which a patient’s or a group of similar patients’ clinical care can be tailored (precisely!).
The discussion surrounding somatic genomic testing at the American Cancer Society’s Cancer Action Network Personalized Medicine Round Table in Philadelphia represents only one of multiple aspects of precision medicine/personalized medicine. Somatic genomic testing is considered standard of care in oncology. As with standards of care, clinical — as opposed to research — somatic genomic testing is often assumed to be a single entity, with uniformly high standards and uniformly high accuracy. Virtually all somatic genomic testing is based on massively parallel sequencing technology, again believed to be identical across clinical labs.
Limitations of testing
However, even this cutting-edge technology has inherent limitations — often known to only too small a handful. One example is the inherent low coverage of certain genes or parts of genes that plagues most platforms. Sequencing coverage also varies from lab to lab in that balance between quality and cost, with some labs using high coverage, whereas others settle for very low coverage.
Variants identified by massively parallel sequencing should always be experimentally validated by another technology, often Sanger sequencing. Unfortunately, not all (clinical) labs validate their initial findings in this manner, or validate at all. Once past the technology, the output from the sequencer requires interpretation. The interpretation of whether an identified variant is pathogenic is what differentiates the quality of labs and is dependent on the databases and clinical outcomes available to a particular lab. Thus, not all (clinical) somatic genomic testing is created equal.
For the patient, a best outcome after somatic genomic testing is to enable the precise choosing of an effective drug, a therapy targeted to the genomic change. However, more than one of the round table panelists pointed out that only 5% of the tumors tested result in treatment changes and selection of genotype-informed targeted treatments. Multiple publications have come out, and more will be published, describing the somatic mutational landscape of diverse malignancies; yet, this 5% figure has not increased in parallel with the growing number of such publications. Perhaps it is time to take a multidimensional approach.
For example, an mTOR inhibitor could be equally effective against a tumor with upregulation of mTOR signaling, whether as a consequence of PTEN mutation or nongenomic means of mTOR upregulation. Might not a multidimensional (multi-omic) analysis include both genomic and nongenomic alterations so that drug targets could be increased?
Germline testing
The elephant in the room of somatic genomic testing is incidentally finding germline (inherited) mutations that predispose to inherited cancer syndromes, which are actionable. From at least five studies on somatic genomic testing, a weighted average of 7% (range, 3-16) had incidental germline mutations in cancer-predisposition genes. Finding germline mutations for hereditary cancers have clinically actionable implications for both patient and family. However, because the majority of clinical somatic genomic testing relies only on testing the tumor, one cannot tell for certain.
It is also the standard of care to ensure genetic counseling before germline predisposition testing. However, it is not practical to offer pretest genetic counseling to every individual whose tumor is undergoing somatic genomic testing, in the context of a limited workforce. A practical and “safe” approach taken by Cleveland Clinic is for the treating oncologist to briefly mention that germline incidental findings are possible before somatic genomic testing, after which, should results suggest the former, the patient would know that he/she has quick access to genetic counseling.
At least one panelist, Vivian H. Coates, MBA, warns that “you cannot simply read the latest one or two journal articles” to be expert and facile at interpreting the somatic genomic testing results, let alone uncover whether incidental germline findings are at play for a particular patient. Hence, the role of somatic genomics tumor boards, which encompass multiple disciplines, is underlined. However, it is surprising that these genomics tumor boards do not rely much more on their physician–scientists who are trained and facile in both worlds of clinical genomics and the underlying basic science.
Further, it is imperative that such genomics tumor boards comprise formally trained physician–scientist clinical cancer geneticists and cancer genetic counselors.
“It is the mark of an instructed mind to rest satisfied with the degree of precision to which the nature of the subject admits and not to seek exactness when only an approximation of the truth is possible.”
— Aristotle
References:
American Cancer Society’s Cancer Action Network Personalized Medicine Round Table; Aug. 11, 2015; Cleveland.
Eng C, et al. “Treating cancer with precision medicine and genetic evaluations.” Cleveland Plain Dealer. Oct. 18, 2015. Available at: www.cleveland.com/opinion/index.ssf/2015/10/treating_cancer_one_gene_at_a.html. Accessed Dec. 21, 2015.
For more information:
Charis Eng, MD, PhD, is professor and chair of the Cleveland Clinic Genomic Medicine Institute and HemOnc Today’s molecular oncology section editor. She can be reached at engc@ccf.org.
Disclosure: Eng reports no relevant financial disclosures.