Abstract 664P
Background
Treatment of RM NPC in non-endemic areas suffers from the lack of prospective trials. We conducted a retrospective multicentre observational study in non-endemic areas in pre-immunotherapy era.
Methods
Clinical data of 1230 patients (pts) from 11 countries with diagnosis of NPC between 2004 and 2018 were collected in a web-based platform (minimum follow up 12 months). Of these, 454 (36.9%) had RM disease. Survival curves were obtained using Kaplan-Meier method and compared with the log-rank test. We used Cox proportional hazards models to test prognostic factors for death and disease progression.
Results
Fifty-eight (13%) pts had metastasis at diagnosis, while 396 (87%) pts relapsed after curative treatment, with a median time to progression (TTP) of 16 months (range 2-236). The table depicts 3-yr PFS and OS in pts subgroups. Patients with loco-regional relapse had a better prognosis (HR 0.48 IC 0.36-0.64). At cox multivariate analysis, younger pts (< 50 years) had a better outcome (HR 0.97 IC 0.95-0.99) as well as who underwent radical treatment as surgery (HR 0.33 IC 0.17-0.66) and/or radiotherapy (HR 0.52 IC 0.28-0.95). RM pts treated with systemic agents received up to 5 lines of chemotherapy (median 3) and the most frequently adopted schedules were platinum-based, in first and second line with an ORR of 68% and 52%, respectively. After starting a first-line chemotherapy 3-yr OS was 34%. Table: 664P
Survival
| Pts subgroups | 3-yr PFS | 3-yr OS |
| Any site of relapse | 22% | 44% |
| Loco-regional relapse exclusively | 27% | 59% |
| Metastatic relapse exclusively | 18% | 31% |
| Loco-regional & metastatic relapse | 10% | 32% |
Conclusions
In non-endemic setting, less than half of the pts with RM NPC are alive at 3 years. Pts treated with surgery and/or radiation for loco-regional relapses show better long-term outcome. Response rate to systemic chemotherapy is quite high and pts may receive sequential treatments, however only a limited group of them obtaining long-term survival. The current data represent the large dataset of RM NPC in non-endemic settings, and they could be considered the benchmark to be compared with new treatment approaches also comprising immunotherapy.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.