A gene expression profiling test predicted worse outcomes for patients with uveal melanoma than actually came to pass, a 10-year study published in JAMA Ophthalmology shows.
Investigators at The Department of Ophthalmology at University of Iowa (UI), Iowa City and Cole Eye Institute, Cleveland Clinic, Ohio reviewed records for 92 and 255 consecutive patients, respectively, (mean age at diagnosis 59.4±13.0 years) .
Investigators at these centers observed the metastasis-free survival (MFS) at their clinics and compared it with a predicted MFS rate developed by a commercially available gene expression profiling (GEP) test
Patients were evaluated with a fine-needle aspiration biopsy GEP. GEP showed class 1A, low risk in 43% of tumors; class 1B, intermediate risk in 22%; and class 2, high risk in 35%. MFS was specified as time-to-metastasis, or censored at the last follow-up date if no metastasis occurred. In total, 48 patients developed metastases — 40 who had class 2 tumors, 5 with class 1A, and 3 with class 1B. The resulting observed 3-year MFS for individuals with class 2 tumors was 67% (95% CI, 59%–77%), and 5-year MFS was 47% (95% CI, 37%–61%).
Investigators also factored in a meta-analysis of published MFS by searching “gene expression profiling and uveal melanoma/choroidal melanoma, prognostication, and uveal/choroidal melanoma” in PubMed. The meta-analysis revealed similar rates to the observed analysis; 3-year MFS for class 2 tumors was 62% (95% CI, 57%–66%), and 5-year MFS 40% (95% CI, 34%–46%).
Both observed and published MFS percentages were higher than genetic prediction. “The predicted MFS for each patient as extracted from the GEP report was 50% and 28%, respectively,” the study notes. Investigators added that predicted MFS fell beyond the 95% confidence interval limits of all a single published study.
Further, tumor variables were analyzed in relation to class 2 cases. According to the Collaborative Ocular Melanoma Study (COMS) criteria for small, medium, and large tumors, patients with metastasis displayed larger tumors than those without metastasis (P =.06), as did thicker lesions (P =.01), greater mean largest basal diameter (LBD) difference (P =.01), and higher American Joint Committee on Cancer (AJCC) stage (P =.04).
Each published study exhibited an amount of data overlap with the observed cohort, except the GEP validation trial which largely established MFS estimation. Lower MFS rates in the trial may have been due to a sample with thicker lesions, more ciliary body tumors, and a larger share of older participants. Researchers speculate integrating tumor size into GEP’s algorithm “will improve on its patient specific prediction.”
The current study was limited by its retrospective nature. A strength involved surveillance imaging of metastases in all cases, and confirmed with biopsy in 99% of patients. Due to the lower GEP MFS rate, a control group is indicated in new trials for adjuvant therapies, the investigators explained, adding “improvement in MFS over the predicted MFS by GEP in a nonrandomized (treatment only) adjuvant therapy trial could be misinterpreted as a therapeutic benefit, particularly in small tumors.”
Disclosures: A study author has declared affiliations with the biotechnology and medical device industries. Please see the original reference for a full list of disclosures.
Singh AD, Binkley EM, Wrenn JM, et al. Predicted vs observed metastasis-free survival in individuals with uveal melanoma. JAMA Ophthalmol. Published online July 21, 2022. doi: 10.1001/jamaophthalmol.2022.2623