Molecular malignancy in melanocytic lesions
Disclosed are methods for determining whether a melanocyte-containing sample (such as a nevus or other pigmented lesion) is benign or a primary melanoma. These methods can include detecting (at the molecular level, e.g., mRNA, miRNA, or protein) the expression of at least two disclosed genes in a biological sample obtained from a subject. Also provided are arrays and kits that can be used with the methods.
1. A method of treating an indeterminant or atypical nevi in a subject, comprising:
measuring nucleic acid expression of
(i) biomarkers MAGEA2, PRAME, PDIA4, NR4A1, PDLIM7, B4GALT1, SAT1, RUNX1, and SOCS3, and
(ii) at least one normalization biomarker(s),
in a nevi biopsy sample obtained from a subject, thereby generating raw expression values for each of the biomarkers and the at least one normalization biomarker(s);
normalizing the raw expression values for each of the biomarkers to the raw expression values for the at least one normalization biomarker(s) to generate normalized expression values for each of the biomarkers;
using the normalized expression values in a regression or machine learning algorithm to generate an output value;
measuring an output value on a same side of a cut-off value as a plurality of known malignant melanoma samples; and
administering one or more chemotherapeutic agents, immunotherapy, or radiation therapy, performing additional surgery to remove lymph nodes or more tissue, or combinations thereof, thereby treating the indeterminant or atypical nevi in the subject.
2. The method of claim 1 , wherein the at least one normalization biomarker(s) has no statistically significant difference in expression between nevi and primary melanoma samples.
3. The method of claim 1 , wherein the at least one normalization biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8 or all 9 of the biomarkers in Table 3.
4. The method of claim 1 , further comprising:
measuring gene expression values for a plurality of biomarkers in the nevi biopsy sample; wherein the range of expression for the plurality of biomarkers is representative of the full range of biomarker expression in the sample transcriptome;
calculating a central tendency expression value for such plurality of biomarkers;
optionally, removing outliers and calculating a recalculated plurality central tendency expression value without the outlier expression values; and
using the central tendency expression value or, optionally, the recalculated plurality central tendency expression value, to normalize the raw expression values for each of the biomarkers.
5. The method of claim 1 , wherein the biomarkers further comprise at least two of the biomarkers in Table 4, and wherein the output value was generated by a logistic regression algorithm.
6. The method of claim 1 , wherein the algorithm is
Output Value=β0+β1 X 1+β2 X 2+ . . . β nXn
wherein Xn are log expression value for the biomarkers MAGEA2, PRAME, PDIA4, NR4A1, PDLIM7, B4GALT1, SAT1, RUNX1, and SOCS3, wherein β0 is greater than −200 and less than 200, wherein all β for n>0 are greater than −1,000 and less than 1,000.
7. The method of claim 1 , wherein the biomarkers further comprise at least two of the biomarkers in Tables 11 and 13, and wherein the output value was generated by a machine learning algorithm.
8. The method of claim 7 , wherein the biomarkers further comprise at least two of the biomarkers in Table 11 or at least two of the biomarkers in Table 13.
9. The method of claim 1 , wherein the biomarkers further comprise:
at least two of BAX, BIRC5, MET, MAGEC2, POLR2J3, ZFYVE16, and BEST1 or
at least two of POLR2J3, BEST1, BIRC5, MET, ZFYVE16, HIF1A, and PICALM, and
wherein the output value was generated by a machine learning algorithm.
10. The method of claim 1 , wherein the biomarkers further comprise at least 50%, at least 75%, at least 80%, at least 90%, at least 95% or at least 98% of the genes in Table 4, Table 11, Table 13, or both Table 4 and Table 11.
11. The method of claim 1 , wherein measuring nucleic acid expression comprises contacting the nevi biopsy sample with a plurality of nucleic acid probes or paired amplification primers, wherein each probe or paired primers is/are specific and complementary to one of the least two biomarkers, under conditions that permit the plurality of nucleic acid probes or paired primers to hybridize to its/their complementary biomarkers.
12. The method of claim 11 , further comprising, after contacting the sample with the plurality of nucleic acid probes, contacting the sample with a nuclease that digests single-stranded nucleic acid molecules.
13. The method of claim 1 , wherein the one or more chemotherapeutic agents comprise dacarbazine, interferon, ipilimumab, carboplatin with taxol, granulocyte macrophage colony stimulating factor (GMCSF), and/or vemurafenib.