Validation of the clinical use of any diagnostic test focuses on 3 main principles: 1) analytic validity of the test; i.e., the technical performance of the test; 2) clinical validity, i.e., the diagnostic performance of the test, such as sensitivity, specificity, and positive and negative predictive values in different populations of patients and compared to the gold standard; and 3) clinical utility of the test, i.e., how the results of the diagnostic test will be used to improve patient management.
Genetic testing typically consists of sequence analysis of the coding regions and intron/exon splice sites or analysis of a specific mutation. Studies report identifying deleterious mutations in the 5' untranslated region and deep intronic mutations in the CDKN2A gene.
The clinical validity is related to the interpretation of the results of the genetic analysis for the individual patient. One issue common to genetic testing for any type of cancer susceptibility is determining the clinical significance of individual mutations. For example, mutations in the CDKN2A gene can occur along its entire length, and some of these mutations represent harmless polymorphisms or noncoding mutations. Interpretation will improve as more data accumulate regarding the clinical significance of individual mutations in families with a known hereditary pattern of melanoma. However, the penetrance of a given mutation will also affect its clinical significance, particularly since the penetrance of CDKN2A mutations may vary with ethnicity and geographic location. (1, 2) For example, exposure to sun and other environmental factors, as well as behavior and ethnicity may contribute to the penetrance. Bishop and colleagues have estimated that the calculated risk of developing melanoma before age 80 years in carriers of CDKN2A mutations ranges from 58% in Europe to 91% in Australia. (6)
Interpretation of a negative test is another issue. CDKN2A mutations are found in less than half of those with strong family history of melanoma. Therefore, additional melanoma predisposition genes are likely to exist, and patients with a strong family history with normal test results must not be falsely reassured that they are not at increased risk. (1) For example, in a 2011 meta-analysis of 145 genome-wide association studies, 8 independent, genetic loci were identified as being associated with a statistically significant risk of cutaneous melanoma, including 6 with strong epidemiological credibility (MC1R, TYR, TYRP1, SLC45A2, ASIP/PIGU/MYH7B, and CDKN2A/MTAP). (7) Also, in a 2011 meta-analysis of 20 studies with data from 25 populations, red hair color variants on the MC1R gene were associated with the highest risk of melanoma but non-red hair color variants were also associated with an increased risk of melanoma. (8) In a 2012 review, Ward and colleagues noted the genetics of melanoma are far from being understood, and “it is likely a large number of SNPs (single nucleotide proteins), each with a small effect and low penetrance, in addition to the small number of large effect, high-penetrance SNPs, are responsible for CMM (cutaneous malignant melanoma) risk.” (9)
In 2009, Yang and colleagues conducted a study to identify modifier genes for CMM in CMM-prone families with or without CDKN2A mutations. (10) The investigators genotyped 537 individuals (107 CMM) from 28 families (19 CDKN2A-positive, 9 CDKN2A-negative) for genes involved in DNA repair, apoptosis, and immune response. Their analyses identified some candidate genes, such as FAS, BCL7A, CASP14, TRAF6, WRN,IL9, IL10RB, TNFSF8, TNFRSF9, and JAK3, that were associated with CMM risk; after correction for multiple comparisons, IL9 remained significant. The effects of some genes were stronger in CDKN2A-positive families (BCL7A and IL9), while some were stronger in CDKN2A-negative families (BCL2L1). The authors concluded that these findings support the hypothesis that common genetic polymorphisms in DNA repair, apoptosis, and immune response pathways may modify the risk of CMM in CMM-prone families, with or without CDKN2A mutations.
In 2010, Kanetsty and colleagues conducted a study to describe associations of MC1R (melanocortin 1 receptor gene) variants and melanoma in a U.S. population and to investigate whether genetic risk is modified by pigmentation characteristics and sun exposure. (11) The study population included melanoma patients (n=960) and controls (n=396), with self-reported phenotypic characteristics and sun exposure information. Logistic regression was used to estimate associations of high- and low-risk MC1R variants and melanoma, overall and within phenotypic and sun exposure groups. Carriage of 2 low-risk, or any high risk MC1R variants, was associated with increased risk of melanoma (odds ratio [OR]: 1.7; 95% confidence interval [CI]: 1.0-2.8; and OR: 2.2; 95% CI: 1.5-3.0, respectively). However, risk was noted to be stronger in or limited to individuals with protective phenotypes and limited sun exposure, such as those who tanned well after repeated sun exposure (OR: 2.4), had dark hair (OR: 2.4), or had dark eyes (OR: 3.2). The authors concluded that these findings indicate MC1R genotypes provide information about melanoma risk in those individuals who would not be identified as high-risk based on their phenotypes or exposures alone. However, how this information impacts patient care and clinical outcomes is not known.
A 2010 article on identifying individuals at high risk for melanoma emphasizes the use of the family history. (12)
While genetic testing for CDKN2A mutations is recognized as an important research tool, its clinical use will depend on how the results of the genetic analysis can be used to improve patient management. Currently, management of patients considered at high risk for malignant melanoma focuses on reduction of sun exposure, use of sunscreens, vigilant cutaneous surveillance of pigmented lesions, and prompt biopsy of suspicious lesions. At present, it is unclear how genetic testing for CDKN2A would alter these management recommendations. The following clinical situations can be considered:
1. Affected individual with a positive family history
If an affected individual tests positive for a CDKN2A mutation, he/she may be at increased risk for a second primary melanoma compared to the general population. However, limited and protected sun exposure and increased surveillance would be recommended to any patient with a malignant melanoma, regardless of the presence of a CDKN2A mutation. However, a positive result will establish a mutation, thus permitting targeted testing for the rest of the family. In addition, a positive mutation in an affected family member increases the likelihood of its clinical significance if detected in another family member. As described, a negative test is not interpretable.
2. Unaffected individual in a high-risk family
If the unaffected individual is the first to be tested in the family (i.e., no affected relative has been previously tested to define the target mutation), it is very difficult to interpret the clinical significance of a mutation, as described. The likelihood of clinical significance is increased if the identified mutation is the same as one reported in other families, although the issue of penetrance is a confounding factor. If the unaffected individual has the same mutation as an affected relative, then the patient is at high risk for melanoma. However, again it is unclear how this would affect the management of the patient. Increased sun protection and surveillance are recommended for any patient in a high-risk family.
The published data on genetic testing of the CDKN2A and CDK4 genes focus on the underlying genetics of hereditary melanoma, identification of mutations in families at high risk of melanoma, and risk of melanoma in those harboring these mutations. Other studies have also focused on the association between CDKN2A and pancreatic cancer. (13-15) One publication added the caution that differences in melanoma risk across geographic regions justify the need for studies in individual countries before counseling should be considered. (16)
In a 2008 study, Aspinwall et al. found short-term change in behavior among a small group of patients without melanoma who were positive for the CDKN2A mutation. (17) In this prospective study of 59 members of a CDKN2A mutation-positive pedigree, behavioral assessments were made at baseline, immediately after CDKN2A test reporting and counseling, and at 1-month follow-up (42 participants). Across multiple measures, test reporting caused CDKN2A mutation carriers without a melanoma history to improve to the level of adherence reported by participants with a melanoma history. CDKN2A-positive participants without a melanoma history reported greater intention to obtain total body skin examinations, increased intentions and adherence to skin self-examination recommendations, and increased number of body sites examined at 1 month.
In a 2011 retrospective case-control study, van der Rhee and colleagues sought to determine whether a surveillance program of families with CDKN2A mutations allowed for earlier identification of melanomas. (18) Characteristics of 40 melanomas identified in 35 unscreened patients (before heredity was diagnosed) were compared to 226 melanomas identified in 92 relatives of those 35 unscreened melanoma patients that were found to have the CDKN2A mutation and participated in a surveillance program over a 25-year period. Surveillance consisted of a minimum of an annual total skin evaluation, which became more frequent if melanoma was diagnosed. Melanomas diagnosed during surveillance were found to have a significantly lower Breslow thickness (median thickness 0.50 mm) than the melanomas identified in the unscreened patients (median thickness 0.98 mm), signifying earlier identification with surveillance. However, only 53% of melanomas identified in the surveillance group were detected on regular screening appointments. Additionally, there was no correlation between length of screening intervals (for intervals less than 24 months) and melanoma tumor thickness at time of diagnosis. The authors also noted that despite understanding the importance of surveillance, patient noncompliance was still observed in the surveillance program, and almost half of patients were noncompliant when first diagnosed with melanoma.
Branstrom and colleagues examined a self-reported survey of genetic testing perceptions and preventive behaviors in 312 family members with increased risk of melanoma. Fifty-three percent had been diagnosed with melanoma, and 12% had a positive susceptibility genetic test. (19) The study indicated that a negative test might be associated with an erroneous perception of lower risk and fewer preventive measures.
Ongoing Clinical Trials
A search of the online site ClinicalTrials.gov identified one observational study, sponsored by the National Cancer Institute, to identify genetic and environmental factors related to melanoma risk in individuals and families at high risk for melanoma (NCT00040352). Another study to develop a model for genetic susceptibility for melanoma is active but no longer recruiting patients (NCT00591500).
Practice Guidelines and Position Statements
The Melanoma Genetics Consortium, comprising familial melanoma researchers from North America, Europe, and Australia, indicated, in 2002, that genetic testing for melanoma susceptibility should not be offered outside of a research setting. (20)
In 2002, in an American Society of Clinical Oncology (ASCO) publication, Kefford noted the sensitivity and specificity of tests for CDKN2A mutations are not fully known. (21) Because interpreting genetic tests is difficult and because test results do not alter patient management, the Kefford publication indicated CDKN2A genetic testing should be performed only in clinical trials for several reasons including: a low likelihood of finding mutations in known melanoma susceptibility genes, uncertainty about the functionality and phenotypic expression of the trait among mutation carriers, and the lack of proven melanoma prevention and surveillance strategies. Additionally, it was noted all patients with risk factors for cutaneous melanoma should follow programs of sun protection and skin surveillance, not just those patients considered to be high risk due to family history.
Because some cases of cutaneous malignant melanoma (CMM) are familial, potential genetic markers for this disease are being evaluated. Some of these markers are being evaluated in those with a family history of disease; other markers are being evaluated to estimate risk of CMM in those who may not have a family history.
The evidence to date is insufficient to permit conclusions concerning the effect of genetic testing for melanoma on health outcomes. While research continues in this area, none of the articles identified demonstrate how the presence or absence of these genetic mutations would impact clinical care—either for those with melanoma or for those at risk due to a family history. The changes in patient management that result from finding a mutation in a patient at risk are not known. In addition, not finding a mutation does not exclude the presence of familial cutaneous malignant melanoma. The conclusion concerning unknown impact on outcomes applies to both mutations with high penetrance (CDKN2A), as well as those with low penetrance (MC1R), which may increase susceptibility. Therefore, genetic testing for mutations associated with familial cutaneous malignant melanoma or associated with susceptibility to cutaneous malignant melanoma is considered experimental, investigational and unproven.