Orbital exenteration and conjunctival melanoma: a 14-year study at the Jules Gonin Eye Hospital.

To report our 14-year experience with orbital exenteration and assess risk factors for poor prognosis by focusing on conjunctival melanoma. A retrospective study was conducted in our tertiary care centre (Jules Gonin Eye Hospital, Lausanne, Switzerland) between 2003 and 2017. Inclusion criteria were patients aged ≥18 years with a follow-up >12 months, without metastatic spread at the time of surgery. Data recorded were age, gender, tumour histology, surgical technique, postoperative complications, surgical margin status, local recurrence, postoperative radiation beam therapy and metastatic status. Twenty-five patients with a mean age of 63.2 years (38-92) were included. Conjunctival melanoma was the most frequently identified tumour (n = 14, 56%) followed by conjunctival squamous cell carcinoma (n = 4, 16%), sebaceous carcinoma (n = 3, 12%), choroidal melanoma (n = 2, 8%) and basal cell carcinoma (n = 2, 8%). Eighteen tumours (72%) originated from the conjunctival tissue. Clear surgical margins were achieved in 21 (84%) patients. Fourteen (56%) patients experienced distant metastases and died from metastatic spread after a mean follow-up of 52.3 months (6-120). The 1-, 3- and 5-year overall survival (OS) was 96%, 72% and 60%, respectively. In the univariate analysis, positive surgical margins, local recurrence and metachronous metastases were associated with a decreased OS (p = 0.002, p = 0.005 and p = 0.007, respectively). In the multivariate analysis, positive surgical margins and metachronous metastases were also associated with a decreased OS (p = 0.02 and p = 0.042, respectively). Conjunctival melanoma was not associated with a poorer prognosis (p = 0.280). Free surgical margins are needed to increase OS. To achieve clearer surgical margins, neoadjuvant targeted therapies/immunotherapies may be considered