Cancer Genetic Counseling—Current Practice and Future Challenges

Abstract

Cancer genetic counseling practice is rapidly evolving, with services being provided in increasingly novel ways. Pretest counseling for cancer patients may be abbreviated from traditional models to cover the elements of informed consent in the broadest of strokes. Genetic testing may be ordered by a cancer genetics professional, oncology provider, or primary care provider. Increasingly, direct-to-consumer testing options are available and utilized by consumers anxious to take control of their genetic health. Finally, genetic information is being used to inform oncology care, from surgical decision-making to selection of chemotherapeutic agent. This review provides an overview of the current and evolving practice of cancer genetic counseling as well as opportunities and challenges for a wide variety of indications in both the adult and pediatric setting.

CANCER GENETIC COUNSELING—WHERE WE HAVE BEEN AND WHERE WE ARE GOING

Historically, individuals with known or suspected risk for an inherited mutation in a cancer predisposition gene were likely to be referred for genetic counseling if their care was received at an academic medical center or Comprehensive Cancer Center. As part of a genetic counseling session lasting ∼1 h, a three-generation pedigree would be constructed and the prior probability that the patient had inherited a germline mutation calculated. Genetic testing was offered for those having an ∼5%–10% chance of having a mutation detected and ordered from a handful of specialty commercial or academic labs. Prior to the 2013 Supreme Court ruling overturning gene patenting, many genetic tests cost $3000 or more, and only the rare patient could afford the cost without meeting a strict set of insurance-based criteria. Genetic counseling and testing, including disclosure of results, was typically provided in person as part of a multivisit, multidisciplinary encounter (Hoskins et al. 1995; Resta et al. 2006).

The practice of genetic counseling has been undergoing rapid change in recent years, as increasing numbers of patients require access to genetic testing for treatment decisions, and broader categories of individuals are now considered appropriate candidates for genetic services. This work reviews traditional models for the practice of cancer genetic counseling, and preliminary evidence supporting the inclusion of novel interventions.

WHO IS A CANDIDATE FOR CANCER GENETIC COUNSELING AND TESTING?

According to SEER estimates from 2014 to 2016, ∼39.3% of men and women will develop cancer in their lifetime, and most will be over the age of 55 yr (https://SEER.cancer.gov/statfacts/html/all.html). Germline pathogenic variants (PVs) in highly penetrant genes play a major role in at least 5%–10% of all cancers and in more than 50 hereditary cancer syndromes. Genetic testing for these syndromes provides cancer risk information that can be used to personalize medical management options to help mitigate the risks. Specific features of a person's medical or family history, or “red flags,” that increase the likelihood of having a hereditary cancer predisposition are summarized in Table 1. Having cancer at an earlier than typical age for the tumor type is a possible indicator of inherited risk and is particularly true for a child diagnosed with an adult-onset tumor, such as colon cancer (Lynch et al. 1979; Cannon-Albright et al. 1989). Certain rare tumors also have a strong association with cancer predisposition. For example, ∼40% of individuals diagnosed with a paraganglioma will have a hereditary PV in one of at least six different genes (Fishbein et al. 2013). An overview of selected pediatric and adult tumors that warrant referral for genetics evaluation regardless of family history are highlighted in Table 2.

Table 1.

Personal and family history “red flags” suggestive of an underlying hereditary cancer predisposition

History	Examples
Personal
Early age at diagnosis	Colon cancer at 30
Rare tumor	Male breast cancer, rhabdoid tumor
Tumor associated with known syndrome	Pheochromocytoma, retinoblastoma
Multifocal or bilateral tumors	Bilateral Wilms tumor
Multiple primary tumors	Breast cancer and ovarian cancer
Lack of known environmental factor	Lung cancer in nonsmoker
Excessive toxicity to treatment	Severe toxicity to chemotherapy in Fanconi anemia
Other developmental and physical differences	Large head size, asymmetry, congenital heart defects, etc. (for comprehensive examples, see Kesserwan et al. 2016; Coury et al. 2018)
Pathogenic variant detected in tumor/abnormal tumor testing	ALK mutation in neuroblastoma, MSH6 absent (IHC) colon cancer
Family
Multiple generations affected with tumors or cancers (on same side of family)
Multiple first-degree relatives affected (parent, child, sibling)
Pattern of cancers suggestive of known syndrome	Osteosarcoma, brain tumor, adrenocortical carcinoma
Known mutation in family member	SDHB mutation in father
Consanguinity	Parents of affected child are first cousins
Ethnicity	Ashkenazi Jewish, Icelandic founder mutations in BRCA1/2

Table 2.

Select tumors that warrant genetics evaluation

Tumor	Highly associated syndromes	Gene	Commonly associated tumor/cancer risks
Adrenocortical carcinoma	Li–Fraumeni syndrome	TP53	Breast, sarcoma, brain tumor, adrenocortical carcinoma, many cancer types
Anaplastic rhabdomyosarcoma	Li–Fraumeni syndrome	TP53	See above
Cerebellar hemangioblastoma	von Hippel–Lindau syndrome	VHL	Endolymphatic sac tumor, pancreatic islet cell carcinoma, hemangioblastoma of CNS or retina, renal cell carcinoma, cysts in the liver, kidney, pancreas, or spleen
Choroid plexus carcinoma	Li–Fraumeni syndrome	TP53	See above
Diffuse gastric cancer	Hereditary diffuse gastric cancer	CDH1	Lobular breast cancer, diffuse gastric cancer
Hypodiploid ALL	Li–Fraumeni syndrome	TP53	See above
Medullary thyroid cancer	MEN2	RET	Medullary thyroid cancer, pheochromocytoma, hyperparathyroidism
Medulloblastoma	Familial adenomatous polyposis	APC	Colon, colon polyps, ampullary, small bowel, small bowel polyps, thyroid, desmoid
	Gorlin syndrome	SUFU, PTCH1	Basal cell carcinoma, ovarian fibroma, jaw keratocyst
	Hereditary breast/ovarian cancer	BRCA2, PALB2	Breast, ovarian, pancreatic, prostate, melanoma
	Li–Fraumeni syndrome	TP53	See above
Ovarian cancer	Hereditary breast/ovarian cancer	BRCA1, BRCA2	Breast, ovarian, pancreatic, prostate
	Lynch syndrome	MLH1, MSH2, MSH6, PMS2, EPCAM	Colon, endometrial, colon polyps, small bowel, urinary tract, ovarian
	Epithelial ovarian cancer	BRIP1, RAD51C, RAD51D	Possibly breast
Ovarian Sertoli–Leydig cell tumor	DICER1 syndrome	DICER1	Pleuropulmonary blastoma, Sertoli–Leydig cell tumor of the ovary, cystic nephroma, thyroid nodules/cancer
Pancreatic cancer	Familial atypical multiple mole melanoma syndrome	CDKN2A	Melanoma, pancreatic, dysplastic nevi
	Hereditary breast/ovarian cancer	BRCA1, BRCA2, PALB2	Breast, ovarian, pancreatic, prostate, melanoma
	Lynch syndrome	MLH1, MSH2, MSH6, PMS2, EPCAM	Colon, endometrial, colon polyps, small bowel, urinary tract, ovarian
	Peutz–Jeghers syndrome	STK11	Breast, gastrointestinal hamartomatous polyps, mucocutaneous pigmentation, colon, lung, small bowel, stomach, cervix, ovarian, testicular
Paraganglioma/pheochromocytoma	Hereditary paraganglioma/ pheochromocytoma syndrome	SDHA, SDHB, SDHC, SDHD, SDHAF2, MAX, TMEM127	Paraganglioma, pheochromocytoma, GIST, renal cell cancer, papillary thyroid cancer
	MEN2	RET
	von Hippel–Lindau syndrome	VHL	Endolymphatic sac tumor, pancreatic islet cell carcinoma, hemangioblastoma of CNS or retina, renal cell carcinoma, cysts in the liver, kidney, pancreas, or spleen
Pleuropulmonary blastoma	DICER1 syndrome	DICER1	See above
Retinoblastoma	Hereditary retinoblastoma	RB1	Melanoma, sarcoma, pineoblastoma
Rhabdoid tumor, atypical teratoid/rhabdoid tumor	Rhabdoid tumor predisposition syndrome	SMARCB1, SMARCA4	Schwannoma, meningioma

(CNS) Central nervous system, (GIST) gastrointestinal stromal tumor.

Any person who develops multifocal, bilateral, or multiple primary tumors has a higher likelihood of having a cancer predisposition (Offit and Brown 1994). However, the likelihood is reduced if the person has had a known environmental exposure such as radiation or certain chemotherapeutic agents to treat a prior cancer. Well-established examples of associations between cancers and environmental exposures include breast cancer after mantle radiation to the chest wall to treat Hodgkin's lymphoma as a teenager (Conway et al. 2017). Certain cancer predisposition syndromes are also associated with significant medical issues, developmental delays, or physical differences, so assessment of additional medical history via targeted questioning or review of medical records may be necessary. For example, individuals with DNA damage repair syndromes such as Fanconi anemia may experience significantly increased toxicity from chemotherapy and display physical manifestations such as multiple congenital anomalies and café au lait spots (de Latour and Soulier 2016). There are many good review articles and guidelines that provide further detail about tumors and nononcologic features of cancer predisposition syndromes and make suggestions for referral and evaluation (Hampel et al. 2015; Kesserwan et al. 2016; Desai et al. 2017; Scollon et al. 2017; Coury et al. 2018; Kennedy and Shimamura 2019).

Determination of who will benefit from genetic testing, tailored screening, and risk-reducing interventions historically began with collection of family history information, and this is still integral to genetics practice today (Pyeritz 2012). Family history and risk factor questionnaires were provided in advance of the consultation to allow patients time to speak with their relatives to optimize the accuracy of the family history. Various software packages are now available to streamline this process. Examples include CRA Health (www.crahealth.com), CancerIQ (www.canceriq.com), Cancer Gene Connect (www.invitae.com/en/cancergeneconnect), and Progeny (Ozanne et al. 2009) (www.progenygenetics.com). Some of these software packages were designed to screen general populations for cancer genetics referral and may also provide quantitative risk assessments. These risk assessment tools are summarized in Table 3.

Table 3.

Risk assessment tools

Tool	Cancer risks assessed	Mutation probabilities	Notable features	Website
Breast Cancer Risk Assessment Tool, aka Gail Model	Breast cancer 5 yr and lifetime risks	None	Used to determine eligibility for chemoprevention Family history and other risk factors	bcrisktool.cancer.gov
Claus Tables	Breast cancer 10 yr and lifetime risks	None	Risk based on family history, including ages at diagnosis	Breast Cancer Risk Assessment Application (BRisk APP)
Tyrer–Cuzick	Breast cancer 10 yr and lifetime risks	BRCA1, BRCA2	Family history and other risk factors including prior negative genetic testing	www.ems-trials.org/riskevaluator
Penn II	None	BRCA1, BRCA2	Based on summary of family history features	pennmodel2.pmacs.upenn.edu/penn2
BOADICEA	Breast and ovarian cancer risks	BRCA1, BRCA2, PALB2, CHEK2, ATM	Family history–based Can calculate cancer risks for those with negative genetic testing results	ccge.medschl.cam.ac.uk/boadicea/boadicea-model
BRCAPRO	Breast and ovarian cancer risks	BRCA1, BRCA2	Family history and other risk factors, includes contralateral breast cancer risk	projects.iq.harvard.edu/bayesmendel/brcapro
MMRpro	Colon and endometrial cancer risks	MLH1, MSH2, MSH6	Risk for Lynch syndrome	projects.iq.harvard.edu/bayesmendel/brcapro
MelaPRO	Melanoma	CDKN2A	Risk for CDKN2A-associated melanoma	projects.iq.harvard.edu/bayesmendel/brcapro
PancPRO	Pancreatic cancer	Dominant pancreatic cancer risk gene	Risk for a putative AD gene for pancreatic cancer risk	projects.iq.harvard.edu/bayesmendel/brcapro
PREMM5	None	MLH1, MSH2, MSH6, PMS2, EPCAM	Based on personal and family history of Lynch-associated cancers	premm.dfci.harvard.edu
ASK2ME	Cancer risks provided in age intervals for mutation carriers	32 high- and moderate-penetrance genes	Risks and summaries of existing management guidelines	ask2me.org/

(AD) Autosomal dominant.

A number of professional societies, including the National Comprehensive Cancer Network (NCCN), have published guidelines to help clinicians determine who should be offered genetic risk assessment and testing. These guidelines, along with additional resources for cancer genetics providers, are summarized in Table 4. As with the risk assessment models, there is variability between professional guidelines. For example, the American Society of Breast Surgeons guidelines recommend genetic testing be offered to all individuals with breast cancer (Plichta et al. 2019), whereas the guideline published by NCCN clearly delineates personal and family history criteria for testing. Insurance companies typically stipulate criteria for coverage of genetic testing, which may differ from professional guidelines. This has led to considerable variation in provider practice, with insurers at times the ultimate arbiters of access based on individual policy coverage decisions.

Table 4.

Cancer and genetics resources and societal guidelines for hereditary cancer predisposition evaluation

Category	Resources	Website	Societal guidelines/policies	Reference	Content summary
Genetics	American College of Medical Genetics and Genomics (ACMG)	www.acmg.net	A practice guideline from the ACMG and the NSGC: referral indications for cancer predisposition assessment	10.1038/gim.2014.147 (Hampel et al. 2015)	Recommendations for referral for cancer genetics evaluation—not guidelines for testing; includes tables of cancer types with personal and family history; features suggestive of specific syndromes; provides description of syndromes and rationale for referral
	National Society of Genetic Counselors (NSGC)	www.nsgc.org
	GeneReviews	www.ncbi.nlm.nih.gov/books/NBK1116/			Comprehensive reviews of genetic conditions
	Genetics Home Reference	ghr.nlm.nih.gov			Information about genes and genetic conditions
General Cancer	American Cancer Society (ACS)	www.cancer.org			Comprehensive information about cancer and support for families
	American Society of Clinical Oncology (ASCO)	www.asco.org	ASCO Policy Statement Update: Genetic and Genomic Testing for Cancer Susceptibility	10.1200/JCO.2015.63.0996 (Robson et al. 2015)	Recommendations for use of multigene panels, somatic testing, and direct-to-consumer testing in clinical care; includes detailed table of informed consent/pretest education
			ASCO policy statement update: genetic and genomic testing for cancer susceptibility	10.1200/JCO.2009.27.0660 (Robson et al. 2010)	Recommendations address inclusion of moderate- and low-penetrance genes and direct-to-consumer testing, not in updated version
	National Comprehensive Cancer Network (NCCN)	www.nccn.org	NCCN Guidelines: Breast/Ovarian, Colon, Gastric, Neuroendocrine and Adrenal Tumors, Myelodysplastic Syndromes	www.nccn.org/professionals/physician_gls/default.aspx	Guidelines for referral for cancer genetics evaluation; includes criteria for genetic testing, criteria for clinical diagnosis, and management guidelines
	National Cancer Institute (NCI)	www.cancer.gov			Comprehensive information about cancer for patients and providers; directory for cancer genetics services
			NCI Cancer Genetics Risk Assessment and Counseling (PDQ)–Health Professional Version. www.cancer.gov/about-cancer/causes-prevention/genetics/risk-assessment-pdq	www.cancer.gov/about-cancer/causes-prevention/genetics/risk-assessment-pdq	Very detailed review of process and content for cancer genetic risk assessment and counseling
	Familial Cancer Database Online (FaCD)	www.facd.info			Database that associates cancer types and syndromes; useful for creating differential diagnosis
Breast/ovarian	The American College of Obstetricians and Gynecologists (ACOG)	www.acog.org	Committee Opinion: Hereditary Cancer Syndromes and Risk Assessment	10.1097/01.AOG.0000466373.71146.51 (2015)	Recommendations for obtaining a family history, medical history, and referral for a cancer genetics evaluation; review of common hereditary syndromes with gynecologic cancers
			Practice Bulletin No 182: Hereditary Breast and Ovarian Cancer Syndrome	10.1097/AOG.0000000000002296 (Committee on Practice Bulletins–Gynecology 2017)	Guidelines for genetic counseling, testing, and management for hereditary breast and ovarian cancer syndrome
	The American Society of Breast Surgeons (ASBrS)	www.breastsurgeons.org	ASBrS: Consensus Guideline on Genetic Testing for Hereditary Breast Cancer	www.breastsurgeons.org/docs/statements/Consensus-Guideline-on-Genetic-Testing-for-Hereditary-Breast-Cancer.pdf	Recommendations for genetic testing for hereditary breast cancer
	Society of Gynecologic Oncology (SGO)	www.sgo.org	SGO statement on risk assessment for inherited gynecologic cancer predispositions	10.1016/j.ygyno.2014.09.009 (Lancaster et al. 2015)	Recommendations for referral criteria for genetic counseling and offering testing for hereditary breast and ovarian cancer and Lynch syndrome
	U.S. Preventive Services Task Force (USPSTF)	www.uspreventive servicestaskforce.org	Draft Recommendation Statement BRCA-Related Cancer: Risk Assessment, Genetic Counseling, and Genetic Testing	www.uspreventiveservicestaskforce.org/Page/Document/draft-recommendation-statement/brca-related-cancer-risk-assessment-genetic-counseling-and-genetic-testing1	Recommendations for risk assessment, genetic counseling, and genetic testing for BRCA1 and BRCA2
Endocrine	American Thyroid Association (ATA)	www.thyroid.org	Revised ATA for the Management of Medullary Thyroid Carcinoma Guidelines	10.1089/thy.2014.0335 (Wells et al. 2015)	Recommendations for RET genetic testing and management of hereditary MTC and MEN2 based on genotype
	Endocrine Society (ENDO)	www.endocrine.org	Pheochromocytoma and Paraganglioma: An Endocrine Society Clinical Practice Guideline	10.1210/jc.2014-1498 (Lenders et al. 2014)	Guidelines for genetic testing and management for hereditary paraganglioma and pheochromocytoma
Gastro-intestinal	American College of Gastroenterology (ACG)	www.gi.org	ACG Clinical Guideline: Genetic Testing and Management of Hereditary Gastrointestinal Cancer Syndromes	10.1038/ajg.2014.435 (Syngal et al. 2015)	Guidelines for genetic testing and management for hereditary gastrointestinal cancer syndromes
	American Gastroenterological Association (AGA)	www.gastro.org	AGA Institute Guideline on the Diagnosis and Management of Lynch Syndrome	10.1053/j.gastro.2015.07.036 (Rubenstein et al. 2015)	Guidelines for diagnosis and management of Lynch syndrome
	American Society of Clinical Oncology (ASCO)	www.asco.org	Hereditary Colorectal Cancer Syndromes: ASCO Clinical Practice Guideline Endorsement of the Familial Risk–Colorectal Cancer: European Society for Medical Oncology Clinical Practice Guidelines	10.1200/JCO.2014.58.1322 (Stoffel et al. 2015)	Guidelines for genetic testing and management for hereditary colon cancer syndromes
			Evaluating Susceptibility to Pancreatic Cancer: ASCO Provisional Clinical Opinion	10.1200/JCO.18.01489 (Stoffel et al. 2019a)	Recommendations for genetic testing and management for hereditary pancreatic cancer syndromes
Hematologic	American Society of Hematology (ASH)	www.hematology.org
Pediatric	American Association for Cancer Research (AACR)	www.aacr.org	Clinical Cancer Research: Pediatric Oncology Series by AACR	clincancerres.aacrjournals.org/pediatricseries	Series of articles summarizing cancer predisposition syndromes in childhood and providing expert consensus guidelines for management
	American Academy of Pediatrics (AAP)	www.aap.org	POLICY STATEMENT: Ethical and Policy Issues in Genetic Testing and Screening of Children	10.1542/peds.2012-3680 (BIOETHICS et al. 2013)	Recommendations about when genetic testing should be offered in different clinical scenarios
	The American Society of Pediatric Hematology/Oncology (ASPHO)	www.aspho.org
	Children's Oncology Group (COG)	www.children soncologygroup.org			Clinical trials group for childhood cancer; provides support for families and connects researchers and clinicians

Further complicating testing and referral decisions are data indicating that personal and family history often fall short in accurately predicting who will test positive for a PV in a clinically actionable gene. Some argue that guidelines, although useful, will miss too many individuals who would test positive (Yurgelun et al. 2015a; Rosenthal et al. 2017; Beitsch et al. 2019). Thus, in many practices the clinical threshold for testing has become less stringent. In fact, some are promoting the merits of population-based screening for cancer predisposition genes (King et al. 2014). Mary-Claire King, PhD, who was the first to localize the BRCA1 gene, has advocated for population-wide screening for BRCA1 and BRCA2, stating that “many women with mutations in these genes are identified as carriers only after their first cancer diagnosis because their family history of cancer was not sufficient to suggest genetic testing. To identify a woman as a carrier only after she develops cancer is a failure of cancer prevention” (King et al. 2014). Others are more measured in their views, but cite population screening as a laudable long-term goal, once the public health impact is better determined (Yurgelun et al. 2015b).

As direct-to-consumer (DTC) genetic testing is becoming more widely available, cancer genetic testing has expanded via this resource as well. Commercial cancer genetic testing companies are offering a patient-driven model of testing, in which a person requests testing online, is paired with a physician or genetic counselor who will order the test, and is offered genetic counseling (McGowan et al. 2014; Covolo et al. 2015). These additional opportunities for consumer-initiated genetic testing will improve access but could also lead to lower testing quality or inappropriate ordering (Phillips et al. 2019). Increasingly, commercial labs are providing bundled services, including ancestry and health information, which may include assessment of cancer risk. One such company, 23andMe, currently offers testing for the three common Ashkenazi Jewish BRCA1 and BRCA2 PVs, as well as analysis of raw data at third-party companies. These approaches have stirred controversy because of concerns that users may not realize this limited genetic testing has little bearing on their chance of having a PV in BRCA1 or BRCA2, unless they are of Ashkenazi Jewish ancestry (Gill et al. 2018). Further concern has been raised about the accuracy of certain DTC tests and third-party interpretations, with one study showing a false-positive result in 40% of samples subsequently sent for clinical confirmation at a CLIA-certified testing lab (Tandy-Connor et al. 2018, 2019).

Cohort studies such as the All of Us Research Program (allofus.nih.gov), The Partners Biobank Registry (biobank.partners.org), and the CSER (Clinical Sequencing Exploratory Research; cser-consortium.org) consortium, are offering genomic sequencing to individuals, unselected for any specific diagnosis, in a variety of clinical settings, and providing results for clinically actionable variants. Initial concerns about these large studies focused on the potential for participants’ adverse psychological reactions and decisional regret after receiving unexpected results. However, so far, results from large-cohort studies have been reassuring, with no reports of clinically significant psychological harm or use of inappropriate medical services from return of results (Hart et al. 2019; Robinson et al. 2019). As access to genetic testing expands, and reassuring data accumulate about psychological outcomes, the number of suitable candidates for cancer genetic services is expected to continue to broaden.

THE PROCESS OF CANCER GENETIC COUNSELING

Historical Context

Genetic counseling is “the process of helping people understand and adapt to the medical, psychological, and familial implications of genetic contributions to disease” (Resta et al. 2006). Genetic counseling was proposed as a key component of the cancer risk assessment process in the 1990s (Peters and Stopfer 1996; Stopfer 2000). Among the first families to receive genetic counseling for cancer risk were those participating in linkage studies aimed at localizing the BRCA1 gene (Biesecker et al. 1993). Empowering patients to make informed decisions regarding screening, prevention, and genetic testing through provision of pertinent genetic and medical information and tailored psychological counseling remain core goals. The Cancer Genetics Studies Consortium (CSGC) Task Force was among the first groups in the oncology setting to develop consensus guidelines for the process and content of informed consent (Geller et al. 1997). This multidisciplinary group first considered why informed consent for genetic testing requires special consideration and acknowledged the following issues: (1) Genetic information affects an entire family, (2) genetic information can present unique challenges for medical professionals as it is probabilistic in nature, and (3) genetic information can lead to the reclassification of patients from healthy to high risk. At that time, the authors described the primary risks and benefits as psychological and social rather than physical or medical because the efficacy of preventive and therapeutic strategies in PV carriers had not yet been substantiated.

In 2004, the National Society of Genetic Counselors published their first set of recommendations for cancer genetic risk assessment and counseling (Trepanier et al. 2004). These guidelines were based on a literature review and the professional expertise of cancer genetic counselors. The Huntington's disease model of pre- and posttest counseling served as an initial template, paying significant attention to the possibility of psychological harm. This model advocated for three separate visits that could each last an hour or more and would sometimes incorporate assessments from mental health professionals before disclosure of results (Biesecker and Garber 1995; Almqvist et al. 2003). Core components of traditional cancer genetic counseling sessions have included (Resta 2006; Madlensky et al. 2017):

• Collection and interpretation of family and medical histories.

• Risk assessment for a cancer diagnosis or for an inherited PV.

• Education about inheritance, testing options, management strategies, resources, prevention, and research opportunities.

• Counseling to facilitate informed decision-making, identification of psychological needs, and provision of appropriate support.

A number of practice guidelines for the provision of cancer genetic counseling and indications for genetic testing in oncology have been published and are summarized in Table 4.

Recent Developments

The evolution of cancer genetic counseling practice has been driven in part by advances in evidence-based medical interventions, social and behavioral research, and expansion of testing options (Athens et al. 2017). The field has moved from a presumption of medical benefit for carriers of cancer susceptibility variants to evidence-based approaches for risk reduction and specialized screening that lead to reduced morbidity and mortality. For example, bilateral salpingo-oophorectomy for women with inherited PVs in BRCA1 or BRCA2 reduces overall mortality, including for those who have already had a breast cancer diagnosis (Domchek et al. 2010; Metcalfe et al. 2015). Patients with Lynch syndrome adhering to screening endoscopy and hysterectomy recommendations demonstrated similar mortality rates to their relatives with negative genetic testing (Järvinen et al. 2009). Testing for RET mutations in families with multiple endocrine neoplasia and age-appropriate prophylactic thyroidectomy based on mutational status improves disease-free survival (Shepet et al. 2013).

Somatic and germline characterization of malignancy is becoming a standard part of the clinical evaluation for an increasing number of oncology patients. Recent studies suggest that nearly 10% of patients with advanced cancer may have actionable PVs that would not have been identified under existing guidelines for clinical testing (Mandelker et al. 2017). Knowledge about the presence of a hereditary PV can influence an oncologist's decision about drug of choice (Goyal et al. 2016; Cohen et al. 2017; Robson et al. 2017; Zhang et al. 2018; Abida et al. 2019; Bast et al. 2019). Therefore, broad categories of cancer patients are now candidates for genetic testing at the time of diagnosis. Because of the relatively high proportion and clinical actionability of PVs found in unselected patients with pancreatic and ovarian cancer, all patients with these diagnoses should be offered germline genetic testing (Stoffel et al. 2019b; Telli et al. 2019).

We now know that information obtained from genetic testing can be lifesaving, and for this reason, genetic testing may appropriately be recommended, rather than simply offered, for certain patients. As expected, when offered a test with implications for cancer treatment, most patients elect to undergo testing (Nilsson et al. 2019). Furthermore, because clinical decision-making will hinge on the outcome of genetic testing, providers and patients are anxious to get the genetic testing process underway and have results available as soon as possible. Therefore, the hesitancy to offer genetic testing in the absence of intensive discussion or a face-to-face encounter has also shifted, as there is a moral imperative to provide equitable access to testing in which a positive result has actionable consequences.

Psychological well-being among those undergoing cancer gene testing has been studied, and the available data do not support initial concerns. Numerous rigorous studies over the years have shown that genetic testing information does not increase the risk for true psychopathology such as suicidality, major depression, or anxiety in someone who does not otherwise carry such a diagnosis (Lerman et al. 1998; Lammens et al. 2010b; Halbert et al. 2011; Kattentidt-Mouravieva et al. 2014). Patients with a prior psychiatric diagnosis, such as a history of depression or who require the use of psychotropic medicines, have demonstrated enhanced psychological vulnerability to genetic testing outcomes in several studies (Murakami et al. 2004; van Oostrom et al. 2007). This highlights the value of assessing prior history of psychiatric diagnoses, and individualizing care when possible, to address psychological support needs when they arise.

Finally, testing options have expanded significantly over the last 10 years, from single-site and single-gene testing to multigene panel tests (Kurian et al. 2018; Phillips et al. 2018). The demand for cancer genetic services has increased because of promising medical management options, lack of serious psychological burden for most individuals, and expansion of laboratory testing options. The need to provide these services to ever-increasing numbers of patients has fostered demand for novel and more efficient models for offering genetic counseling and testing (Hoskovec et al. 2018; McCuaig et al. 2018).

Pretest Counseling Models

The emergence of new models for the provision of pretest counseling challenges providers to ensure the core goals and values of genetic counseling are maintained. A major shift in clinical practice affecting the informed consent discussion has been the transition to multigene panel testing. The “tiered-binned model” of genetic counseling and informed consent was developed to address the content of a genetic counseling sessions for this purpose (Bradbury et al. 2015, 2016). Traditional models often included a detailed discussion about the clinical manifestations and management options of each gene tested. The tiered-binned model provides a layered approach, with broad concepts and common denominator elements required for all patients considered tier 1 “indispensable” information. More specific tier 2 information is provided as needed, to support the informational needs and specific preferences among diverse patient populations. “Binning” refers to the approach of organizing discussion about genes into categories, defined by variables such as high or moderate risk genes, or based on clinical utility (Bunnik et al. 2013).

Alternative pretest service delivery models can increase access to genetic testing by minimizing the number of required in-person appointments and reducing waiting times for results while maintaining patient satisfaction (McCuaig et al. 2018). Alternative models may be led by a genetic counselor, physician, or other provider (Schwartz et al. 2014; Senter et al. 2017; Colombo et al. 2018). Some patients are being offered direct access to genetic testing with minimal to no pretest discussion, with pretest education offered via a video, Web-based education, written information, a chatbot, or interactive relational agent (Sie et al. 2014; McCuaig et al. 2019). These models are summarized in Table 5. Typically, these alternative service delivery models rely on the availability of genetic counseling and further consultation for those found to have a PV. A critical part of posttest counseling will continue to be facilitation of cascade testing for individuals with hereditary risk, as identifying family members with PVs can lead to targeted cancer prevention for these at-risk relatives (Caswell-Jin et al. 2019).

Table 5.

Alternative service delivery models for genetic counseling and testing

	Method	Potential benefits	Potential drawbacks
Genetic counselor–driven	Telephone	No travel needed, widely available interface	Limited reimbursement, no visual cues in the counseling session
	Videoconference	No travel needed, visual cues, may feel more personalized	Technical difficulties more common and may not be billable
	Group counseling	Group setting allows learning from other's questions, may feel supported	Lack of privacy, scheduling opportunities may be limited
	Embedded genetic counseling	Access to in-person counseling same day receiving treatment, increased likelihood of referral	Counseling may not be private if done in infusion area
Nongenetic counselor–driven	Tumor testing flags possible germline	Automated, enriches population of patients likely to have germline mutations	Tumor-only testing will miss detectable germline mutations
	Direct genetic testing	Rapid access	Unexpected results and lack of understanding
	Direct to consumer	Increases patient autonomy and privacy	Unexpected results and no relationship with local provider

Although these models have been found noninferior to in-person genetic counseling in areas such as satisfaction, distress, and knowledge about genetic testing, many patients have preferred in-person genetic counseling (Schwartz et al. 2014; Sie et al. 2014). In addition, it is important to note that the physician-led approaches may not be generalizable to all settings, as many physicians are not comfortable ordering and interpreting genetic test results (Eccles et al. 2015; Kurian et al. 2018). Most studies published to date were conducted as research initiatives within academic medical centers and have focused heavily on more common indications such as hereditary breast and ovarian cancer and Lynch syndrome, potentially limiting the generalizability of conclusions to community-based settings (Manchanda et al. 2016; Bednar et al. 2017; Tutty et al. 2019).

As evidence-based novel models of pretest counseling and education are implemented, care should be taken to identify patients who are undecided about genetic testing after a brief intervention or who prefer a more detailed or in-person session because of psychological or informational concerns. If a genetic counselor is not available locally, remote access via telegenetic services is an option. Like so much of the practice of genetic counseling, using a flexible approach tailored to the patient facilitates optimal care, balancing efficiency with attention to specific psychological needs and preferences.

LABORATORY, TEST SELECTION, AND INTERPRETATION ISSUES

Germline genetic testing has been a valuable tool to elucidate inherited susceptibility to cancer for several decades (Ponder 1994). Genetic counselors are often at the forefront of test selection and are tasked with ensuring individuals make an informed choice about testing. The expansion of testing options has led to increased debates within the field and variability in practice.

Biosample Selection

It is important to understand the characteristics of the biological sample used for testing and its limitations. Blood or saliva is commonly used for germline genetic testing. However, if the patient has a hematologic malignancy, the test results may reflect somatic PV in the cancer. A somatic-only PV could be misinterpreted as germline, or a germline mutation could be missed. Similarly, a PV may be detected in TP53 or another gene at a <0.50 variant allelic fraction in a person without a known hematologic malignancy and may represent a somatic change in the blood with unknown significance, referred to as clonal hematopoiesis of indeterminant potential or aberrant clonal expansions (Steensma 2018; Weitzel et al. 2018). A result like this may be misinterpreted as constitutional somatic mosaicism. In these cases, DNA testing from cultured skin fibroblasts is recommended, and eyebrow plucks are another potential source of DNA. A cheek swab or saliva sample may be contaminated with white blood cells and should be avoided if possible (Weitzel et al. 2018).

Laboratory Selection

An equally important aspect of the hereditary cancer evaluation is the selection of a laboratory. Whether single-gene analysis or a large multigene panel is indicated, it is critical to consider the capabilities and expertise of the lab performing the test. Even if the same basic technology is used, the amount of coverage, such as inclusion of deep intronic PVs, rearrangements, or pseudogene regions, may vary and significantly impact variant detection rates. Ensuring the gene or condition highest on the patient differential is thoroughly assessed provides the best sensitivity and detection rate (Richards et al. 2015).

Another key consideration is the laboratory's experience with interpreting germline variants within cancer genes. Variant interpretation incorporates subjective assessment of available data and may vary depending on a lab's experience (Amendola et al. 2016; Kim et al. 2019). The lack of diversity of those tested in large laboratory cohorts and within publicly accessible databases can lead to errors in classification and interpretation of findings (Manrai et al. 2016). Increasing representation of minority communities in testing cohorts will eventually lead to more accurate test interpretation and fewer variants of uncertain significance (VUSs); however, this will take time.

Test Selection

As with any new development, there is excitement, fear, reservation, and controversy surrounding multigene panel testing. Health-care providers typically seek interventions that maximize benefits while minimizing harm to the patient. Test selection and panel size are vigorously debated in the cancer genetics professional community (Lynce and Isaacs 2016; Lee et al. 2019b). On one extreme, there are those who believe a minimal amount of testing is most appropriate, such as single-site testing for a person with a known family history of a PV. On the other hand, some support the use of broad multigene panel testing that includes high and moderate penetrance genes for which the patient may or may not be at risk based on personal and family history (Douglas et al. 1999; Beadles et al. 2014; Desmond et al. 2015; Tung et al. 2015; Zhang et al. 2015; Parsons et al. 2016).

Even those who use multigene panel testing struggle with determining the optimal panel size and the degree to which a provider should steer patient choice for test selection (Domchek et al. 2013). It may not be appropriate to task patients with this burden when they are in crisis or under significant stress because of a new cancer diagnosis or family history concern. A potential middle ground may be to generally offer a broad multigene panel and help the patient decide if a smaller test is better for them based on their values, beliefs, and goals.

Germline Genetic Testing beyond Panels

Additional germline testing options are on the horizon. Exome and genome sequencing are becoming more widely available and may be helpful if the cancer patient also has congenital anomalies or developmental delay (Diets et al. 2018) or when multigene cancer panels are negative (Powis et al. 2018). Other technologies available include RNA sequencing paired with DNA analysis that can identify the presence and quantity of RNA transcripts, which may help in classification of VUS results and identify deep intronic PVs (Farber-Katz et al. 2018). We can expect that these and other enhancements to laboratory testing techniques will continue to increase the sensitivity of variant detection and clarity of interpretation over time.

Polygenic Risk Scores

Polygenic risk scores (PRSs) incorporate the combined effects of single-nucleotide polymorphisms (SNPs) identified from large-scale genome-wide association studies (GWASs) to estimate a person's risk to develop cancer. Generally, the magnitude of association of each individual SNP with disease risk is small (e.g., conferring a relative risk of <2). There is hope that the cumulative effect of 100 or more informative SNPs on their own, or in combination with a risk model such as Tyrer-Cuzick, can be used to stratify individuals into higher and lower risk categories that may lead to differential screening recommendations. In addition, models such as BOADICEA are working to incorporate SNP data to further refine lifetime risk estimates for those who carry PVs in known cancer susceptibility genes (Lee et al. 2019a). However, many PRSs to date contribute relatively small improvements in risk prediction after other known risk factors are considered. In addition, commercial laboratories are offering SNP-based PRS genetic testing before clinical utility has been proven. Finally, there is differential access to this risk information, because GWAS data are less available for populations other than those of European ancestry, potentially creating or exacerbating disparities in health care. Efforts to study more diverse populations are needed to determine clinical utility and achieve equity in this area of personalized medicine (Martin et al. 2019).

Integration and Interpretation of Genetic Test Results

Cancer risk management recommendations should be based on the integration of the genetic test results with the patient's personal and family history (Robson et al. 2015). For example, a 40-yr-old woman who had breast cancer and has a family history meeting clinical criteria for Li–Fraumeni syndrome (LFS) would receive LFS guidelines even if her testing was negative or identified a VUS in TP53, whereas a similar woman without supporting family history would not receive such guidelines. Posttest risk integration is essential for any streamlined counseling and testing model to ensure that high-risk patients, like the former, are not not missed.

A continuing challenge for cancer genetics providers is managing VUS results. Approximately 90% of VUSs are later reclassified as benign, whereas a much smaller proportion are upgraded to pathogenic (Slavin et al. 2019; So et al. 2019; Tsai et al. 2019). Unfortunately, there is evidence (Kurian et al. 2018) that some providers are making inappropriate surgical or surveillance recommendations based on a lack of understanding of a VUS result, which are common in multigene panels (LaDuca et al. 2014; Kapoor et al. 2015).

It can also be challenging to interpret the significance of a PV in a less well-characterized gene, because cancer risks may not be clearly delineated and consensus management guidelines for follow-up may be lacking. In addition, when multigene panel testing is used, a challenge arises when PVs are found in genes considered highly penetrant, such as TP53 or CDH1, for patients without suggestive personal or family history (Lynce and Isaacs 2016). These findings challenge penetrance estimates and decisions regarding appropriate medical management (Rana et al. 2019; Xicola et al. 2019).

SPECIAL ISSUES IN PEDIATRIC ONCOLOGY

Pediatric cancer genetics is an emerging subspecialty within cancer genetics that combines traditional cancer genetic counseling and pediatric genetics. There are unique aspects to pediatric oncology counseling that result in further complexity in the sessions.

When gathering a medical history for a child who has had cancer, it is important to complete a birth history, full medical review of systems, and a dysmorphology exam if possible, because increased cancer risk could be part of a broader Mendelian syndrome (Kesserwan et al. 2016). As is seen in the adult cancer clinic, there is a trend toward increased use of multigene panel testing, with debate over panel size. However, genetic testing in children raises additional ethical considerations. Genetic testing is typically offered when there is potential medical benefit to the child, such as a clear increased risk for pediatric cancer for which screening is available, along with the expectation of minimal harm (Committee on Bioethics et al. 2013).

Use of an expanded cancer panel in the pediatric population that includes adult-onset cancer genes is controversial. The identification of a PV in a gene known only to be associated with adult-onset tumors may or may not be the cause of a pediatric malignancy (Sylvester et al. 2018). As there are limited medical benefits to learning a child has a PV for an adult-onset cancer and potential harm (lack of future autonomy and potential psychological harm), children are not typically offered testing for adult-onset cancer predisposition genes, such as BRCA1, BRCA2, or Lynch syndrome (Kesserwan et al. 2016). However, this perspective is being challenged, and the testing approach is evolving (Druker et al. 2017; Sylvester et al. 2018; Kuhlen et al. 2019). Although not currently the standard practice in pediatrics, it is possible that knowing germline status of such genes may impact treatment opportunities, especially in well-advanced cancers. Tumor testing is increasingly being performed that may provide insight into the germline, and the information could benefit other family members. In addition, exome analysis is routinely performed in clinical genetics programs, with genetic variants reported as secondary findings according to the recommended ACMG list of actionable genes (Kalia et al. 2017). There are multiple large-panel tests and exome sequencing research studies ongoing that will assess the clinical and psychological impact of this testing in pediatric oncology.

It can be very challenging to involve a child in the pre- and posttest discussions, because parents can have strong opinions about what information should be shared and when. The benefit of open communication with children has been demonstrated in multiple studies, which show enhanced coping and adaptation, decreased anxiety, and increased family cohesion (Rowland and Metcalfe 2013; Valdez et al. 2018). Providers can offer to meet with parents before bringing the child into the visit to allow a more open discussion about the risk assessment and testing options, obtain full consent, and make a plan for how the child will be involved. Assent for genetic testing is recommended when possible (Committee on Bioethics et al. 2013), and some institutions have policies regarding the age of assent, which is commonly around age 11 yr. The amount of information and level of detail provided in these discussions will vary based on the developmental age of the child (Werner-Lin et al. 2018).

Management of pediatric cancer risk can be challenging because of the paucity of empiric data on surveillance or prevention in childhood. Existing guidelines are mostly based on expert consensus opinion (Brodeur et al. 2017). Studies have shown that parents often have an interest in pursuing genetic testing that could lead to enhanced medical management for their child or family (Lammens et al. 2010a; Rasmussen et al. 2010; Alderfer et al. 2015; McCullough et al. 2016; Brozou et al. 2018; Desrosiers et al. 2019). However, there are limited data on the emotional and financial burden that families face with ongoing surveillance and future studies are needed (Weber et al. 2019).

For the parents of a child with cancer, there is a high risk for psychological distress due to the cancer diagnosis (Pai et al. 2007), and the possibility of a hereditary component can be an additionally overwhelming (Lammens et al. 2010a). Parents fear for their child's future, worry about other children or parents, and may feel guilty or responsible for their child's diagnosis, but the benefits of testing often outweigh the potential downsides for parents who choose testing (McCullough et al. 2016; McGill et al. 2019; Scollon et al. 2019).

CONCLUSION

Cancer genetic services are evolving as they become more fully integrated into oncology care. In the past, these services were primarily provided at academic medical centers for those meeting specific personal and family history criteria. Testing to assess cancer risk is offered more broadly now through incorporation of novel service delivery models to reach increasingly diverse adult and pediatric patient populations. Opportunities to reduce morbidity and mortality for those with inherited cancer risk, as well as to stratify targeted therapy for cancer patients, requires equitable and appropriate access. Whereas numerous robust studies have provided promising results about the psychological well-being of those who have had genetic testing for inherited cancer risk, further studies will be critical as this testing extends to more diverse patient populations tested with genomic sequencing, who may have received only a brief educational intervention prior to testing. Genetic counselors will continue to facilitate interpretation of complex test results for both patients and providers and will remain critical to the delivery of high-quality, compassionate genetic services. Those with identifiable gene variants or significant family histories will continue to benefit from genetic counseling practices that help people understand and adapt to the clinical, psychological, and familial implications of genetic contributions to their disease.