Skip to main content
  • Metastasis Risk After Biopsy for Posterior Uveal Melanoma

    By Lynda Seminara
    Selected By: Stephen D. McLeod, MD

    Journal Highlights

    Ophthalmology, December 2018

    Download PDF

    Do the risks of biopsy in patients with uveal melanoma outweigh the benefits? Although some investigators have found a low frequency of ocular complications after such biopsies, the long-term risks have not been studied extensively. In a large longitudinal study spanning 32 years, Bagger et al. looked at the risk of metastasis after biopsy for posterior uveal melanoma. They found that rates of all-cause and melanoma-specific mortality were similar between biop­sied and nonbiopsied patients.

    This study included all patients with posterior uveal melanoma treated in Denmark between January 1985 and December 2016 (N = 1,637). Clinical and histopathologic findings for the study population were linked to pathol­ogy, cancer, and mortality registries. Patients had follow-up from diagnosis of choroidal or ciliary body melanoma until migration, death, or study conclu­sion (November 2017). Data included age, sex, tumor characteristics, and diagnostic and therapeutic measures.

    The absolute risk of melanoma-specific death was denoted by cumu­lative incidence curves that accounted for competing risks. Cox regression models were applied to estimate crude and adjusted hazard ratios and 95% confidence intervals for all-cause mor­tality and melanoma-specific mortality among patients and to com­pare data between biopsied and nonbiopsied cohorts. Fine and Gray risk regres­sion served as a sensitivity analysis of the effects of competing risks.

    Of the 1,637 patients, 567 (35%) had a biopsy during primary treatment. At the time of diagnosis, those who received a biopsy had better prognostic factors, includ­ing smaller tumor size and younger age. Adjusted analyses showed no meaningful differences between the study groups in all-cause mortality or melanoma-specific mortality.

    The original article can be found here.