By Abby Van Voorhees, MD, March 02, 2015
In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with Boris Bastian, MD, PhD, and Jeff North, MD, about their recent American Journal of Surgical Pathology article, "Fluorescence in situ hybridization as an ancillary tool in the diagnosis of ambiguous melanocytic neoplasms: a review of 804 cases."
Dr. Van Voorhees: Lesions suspicious for melanoma are generally excised. How often are these lesions challenging to diagnose?
Drs. Bastian and North: The majority of pigmented lesions biopsied to assess for melanoma can be reliably diagnosed with routine histopathologic assessment by a qualified dermatopathologist without any additional testing. Given the study data on discordance rates among pathologists for melanocytic tumors combined with questioning of some fellow dermatopathologists, a rough approximation is between 5-15 percent of melanocytic neoplasms are challenging to diagnose. The percentage is dependent on the patient population and pathologist level of training. The percentage will likely be higher for patients with extensive sun damage and patients with numerous clinically atypical nevi (e.g. atypical mole syndrome/dysplastic nevus syndrome), and higher with pathologists that are not specialized in dermatopathology.
Dr. Van Voorhees: What are the reasons?
Drs. Bastian and North: Some melanocytic tumors have conflicting histopathologic features, showing some findings typically associated with melanoma (e.g. pagetoid scatter, poor circumscription, irregular distribution of melanocytes) and some associated with melanocytic nevi (e.g. dermal maturation, lack of pleomorphism, minimal mitotic activity). This can lead to diagnostic uncertainty. Another potential problem arises when partial biopsies are submitted and the entire neoplasm cannot be assessed. When both of those occur in the same biopsy, the chance for misdiagnosis can be high.
Dr. Van Voorhees: Do we need additional tests that increase the accuracy of our ability to diagnose melanomas?
Drs. Bastian and North: Most definitely. Studies of discordance show that even expert dermatopathologists disagree in difficult melanocytic tumors. The tendency of many dermatopathologists will be to skew the diagnosis of ambiguous melanocytic tumors toward melanoma in order to prevent undertreatment of a potentially lethal tumor and minimize medical-legal risk for the doctors involved. However, this also results in overtreatment of some patients and undue stress for the patient given a diagnosis of melanoma that may be incorrect. Studies showing rising melanoma incidence proportional to the increase in skin biopsies performed without a corresponding change in melanoma mortality suggest that overdiagnosis of melanoma has contributed to the rise in melanoma incidence. Additional tests to assist in the diagnosis of melanocytic tumors are needed to help increase diagnostic accuracy.
Dr. Van Voorhees: In case some of our readers don’t recall what the FISH tool is, can you remind us what it is and how it works as a diagnostic tool?
Drs. Bastian and North: Fluorescent in situ hybridization (FISH) uses fluorescently labeled nucleic acid probes to assess for chromosomal copy number changes. When the probes are hybridized to the DNA of the neoplastic cell population, the copy number is assessed by counting the fluorescent dots in the nucleus, with each dot representing one copy of the chromosomal region to which the probe was targeted. Prior studies using comparative genomic hybridization (CGH) have shown that genomic instability in the form of chromosomal gains and losses is detectable in >95 percent of melanomas, but is not seen in melanocytic nevi (with the exception of a small subset of Spitz nevi). The chromosomal gains and losses in melanomas are not random and occur in areas of the genome where oncogenes and tumor suppressor genes are found. Based on CGH data, a combination of four FISH probes targeting loci on chromosomes 6 and 11 was shown to distinguish between melanomas and melanocytic nevi with 87 percent sensitivity and 95 percent specificity. There is some indication that the sensitivity is lower for Spitzoid melanocytic tumors, and a newer probe set has been developed that has a reportedly higher sensitivity and specificity.[pagebreak]
Dr. Van Voorhees: In your study how useful was FISH in aiding in the diagnoses of these ambiguous pigmented lesions?
Drs. Bastian and North: FISH is a very useful adjunctive test. Of the 754 tumors tested in our study, FISH testing resulted in a more definitive diagnosis of benign or malignant in 88 percent of cases. The remaining 12 percent were given an equivocal final diagnosis despite FISH testing, almost all of which tested negative with FISH. Thus, a negative FISH result in an ambiguous melanocytic tumor is somewhat reassuring, but does not fully exclude the possibility of melanoma.
Dr. Van Voorhees: How accurate is FISH?
Drs. Bastian and North: This question is difficult to answer, as it will require comparison to a reliable gold standard, which does not exist. Outcome information such as melanoma metastasis or death are definitive endpoints, but not all patients with truly malignant tumors will reach those, as their spread may have been prevented by the excision.
Dr. Van Voorhees: What percent of FISH-positive cases are melanoma? Does this differ across different types of melanoma?
Drs. Bastian and North: Around 95 percent overall. Aberrations of chromosome 6 tested by three of the probes are common across all types of melanoma, including uveal melanoma. However, copy number increases of the fourth probe that targets CCND1 on chromosome 11q13 are significantly more common in acral melanoma and melanomas on skin with chronic sun damage. The probes included in the FISH panel were validated for testing in all types of melanoma with reported sensitivity of around 85 percent, with the exception of desmoplastic melanoma. The limited data available shows the sensitivity of FISH in desmoplastic melanoma is around 50 percent.
Dr. Van Voorhees: Is there value in a negative FISH test even if the lesion turns out to be a melanoma?
Drs. Bastian and North: While around 10-15 percent of melanomas will have a negative FISH test, FISH has shown prognostic value in predicting metastasis in melanomas. Thus a negative FISH test does have some value even if the tumor actually is a melanoma (false negative FISH test), as a FISH-negative melanoma would have a higher chance of behaving more indolently.
Dr. Van Voorhees: In your experience, how often does FISH fail to be helpful in these ambiguous types of lesions?
Drs. Bastian and North: In around 5 percent of FISH cases there is either insufficient tissue remaining or poor hybridization signals that make FISH analysis impossible. In the remaining 95 percent, FISH provides at least some useful diagnostic information about the tumor. Given the high specificity of FISH, a positive test that supports a diagnosis of melanoma is more helpful than a negative FISH test. As not all melanomas have chromosomal aberrations in the regions of the genome assessed by FISH, false negatives are the greatest concern.
Dr. Van Voorhees: Have you identified any particular challenges in performing this type of testing that dermatologists and dermatopathologists should be aware of?
Drs. Bastian and North: Any dermatologist or dermatopathologist who uses FISH should know the limitations of FISH, particularly that both false negative and false positive results can occur. The test should be performed by someone who has experience with performing FISH on melanocytic tumors. It is critical to assess a random sample of nuclei within an area of the tumor that is suspicious to harbor copy number changes and NOT cherry-pick’ individual nuclei based on their probe count. This approach avoids observer bias. Additionally, pathologists should exercise caution when cutting level sections into tissue blocks of melanocytic tumors. Too many level sections or immunostains can exhaust the tumor tissue in the biopsy so that FISH cannot be performed.[pagebreak]
Dr. Van Voorhees: Do you feel that this type of testing should be a part of routine pigmented lesion interpretation, or limited to ambiguous lesions only?
Drs. Bastian and North: The utility of FISH lies in the assessment of melanocytic tumors with ambiguous histopathologic features. The only additional use is for unequivocal melanomas with an intradermal nevoid component that is ambiguous. In this setting FISH can increase the accuracy of microstaging with depth measurement. While there is some prognostic value in the results of FISH, the cost of FISH should limit its use to these select cases.
Dr. Van Voorhees: What other tests are available to assist in the diagnosis of melanocytic tumors?
Drs. Bastian and North: In addition to FISH, CGH is now clinically available to assess for melanoma-associated chromosomal aberrations in ambiguous melanocytic tumors. CGH analysis involves extracting tumor DNA from a routine biopsy, fluorescently labeling that DNA, and then hybridizing it onto a microarray with representative portions of the entire genome. This allows for assessment of chromosomal gains and losses across the entire genome instead of a few small loci that FISH targets. Both FISH and CGH have limitations and can be used in a complimentary way to optimize molecular testing in the diagnosis of difficult melanocytic tumors.
Dr. Bastian is professor of dermatology and pathology at the University of California San Francisco School of Medicine. Dr. North is assistant professor of dermatology and pathology at the University of California San Francisco School of Medicine. Their article appeared in the June 2014 issue of the American Journal of Surgical Pathology. Am J Surg Path 2014 Jun 38(6):824-31. doi:10.1097/PAS.0000000000000189.