Circulating cell free DNA to predict recurrence in uveal melanoma.

Melanoma/Skin Cancers
Session Type and Session Title: 
Poster Session, Melanoma/Skin Cancers
Abstract Number: 


Poster Board Number: 
Board #174
J Clin Oncol 34, 2016 (suppl; abstr 9569)
Ryan Michael Weight, Shingo Sato, Marlana M. Orloff, Michael J. Mastrangelo, Takami Sato; Thomas Jefferson University Hospital, Philadelphia, PA; Thomas Jefferson University, Philadelphia, PA; Department of Medical Oncology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA

Abstract Disclosures


Background: Uveal melanoma is the most common primary intraocular malignant tumor in adults. Up to 50% of uveal melanoma patients die of metastasis, usually to the liver. Cell free DNA (cfDNA) provides a commercially available, non-invasive mechanism for monitoring disease activity in malignancy. Recently, studies have investigated cfDNA in uveal melanoma and have demonstrated an association with hepatic metastasis, metastatic volume as well as PFS and OS. It remains unclear if cfDNA can predict recurrent metastatic disease in high-risk uveal melanoma patients. Methods: We conductedan exploratory study of cfDNA following treatment for primary intraocular uveal melanoma. Three cohorts were investigated: high-risk patients ( > 50% estimated recurrence rate) with no clinical evidence of metastatic disease (cohort 1), patients with newly developed metastatic disease of the liver by surveillance MRI (cohort 2), patients with previously established metastatic disease (cohort 3). cfDNA profiles were provided by Guardant360 complete exon sequencing and analyzed for the presence of GNAQ/GNA11 mutations and MYC amplification which represent driver mutations present in > 80% of uveal melanomas. cfDNA was evaluated within 1 week of surveillance imaging studies. Results: Cohort 1 revealed no G-protein/MYC abnormalities by cfDNA (0/32). 34.4% (11/32) of patients in cohort 1 had mutations unrelated to G-protein/MYC. Cohort 2 showed a 10% (1/10) G-protein/MYC detection rate. Cohort 3 showed a G-protein MYC detection rate of 75% (18/24) which approaches the expected mutation rate in uveal melanoma. Interestingly, cfDNA did not detect G-protein/MYC abnormalities in hepatic tumors less than 2.0cm in diameter, regardless of cohort. Conclusions: MRI is a more sensitive screening test than cfDNA for detection of hepatic metastasis in uveal melanoma. cfDNA negative for G-protein/MYC alterations correlates with the absence of clinically detectable disease (100% specificity) and a positive result in the adjuvant setting should prompt further evaluation. This study suggests a detection threshold for cfDNA of > 2.0cm hepatic tumor diameter, which may have implications for the utilization of cfDNA across tumor types.