The benefit of adjuvant CT in NPC is controversial. We conducted a biomarker guided RCT using post-RT EBV DNA to select high risk NPC pts for adjuvant CT while sparing low risk pts from unnecessary toxicity.
Eligible NPC pts of AJCC (6th Ed) stage IIB-IVB, ECOG 0-1, adequate organ functions, no locoregional disease or distant metastasis after RT/CRT, were screened for plasma EBV DNA at 6-8 weeks post-RT. Pts with EBV DNA =0 (copies/ml) were observed. Pts with EBV DNA >0 underwent work-up and were randomized to arm A (adjuvant cisplatin-gemcitabine x 6 cycles) or arm B (observation). Primary endpoint for the RCT was relapse free survival (RFS). With a hazard ratio (HR) of 2, 100 randomized pts were required for a power of 0.8 and an alpha at 0.05.
From 9/2006 to 7/2015, 789 pts were screened by plasma EBV DNA: 573 (72.6%) pts had EBV DNA =0, 216 (27.4%) pts had EBV DNA >0, 104 (13.2%) pts were randomized (arm A: 52; arm B: 52). The clinical characteristics and significant biomarkers (post-RT EBV DNA levels: 1-49/50-499/>=500; PET-CT scan: negative/positive) were evenly balanced in the two arms. After median follow up of 6.6 years (yr), there was no significant difference in RFS or overall survival (OS) between the two arms in the intent-to-treat and per protocol population, and by subgroup analysis defined by EBV DNA level and PET-CT. By recursive partitioning analysis (RPA), we identified three prognostic groups among 789 post-RT NPC pts (Table). The low risk group (EBV DNA < 50 and stage II/III) shared the same survival (5-ys OS 89%) as stage II but included more than twice the number of pts.Table: 336O
|Risk groups||N (% of total population, n = 789)||No. of events||3-yr OS rate (95% C.I.)||5-ys OS rate (95% C.I.)||HR (95% C.I.)||P value|
|(1) Stage II||220 (27.9%)||36||92.5 (88.1-95.4)||88.2 (82.9-92.0)||1.00|
|(2) Stage III||374 (47.4%)||72||88.3 (84.4-91.2)||81.2 (76.3-85.2)||1.45 (0.97-2.17)||P = 0.069|
|(3) Stage IV||195 (24.7%)||73||79.1 (72.5-84.3)||67.3 (59.5-73.9)||3.08 (2.06-4.59)||P 0||216 (27.4%)||89||70.8 (64.1-76.5)||60.2 (52.9-66.8)||3.18 (2.38-4.27)||P |
Adjuvant CT with cisplatin-gemcitabine did not improve survival in high risk NPC pts with detectable post-RT plasma EBV DNA. The low risk group (includes 66% of pts) can potentially be spared the toxicity of adjuvant CT.
Clinical trial identification
Legal entity responsible for the study
Comprehensive Cancer Trials Unit, Department of Clinical Oncology, The Chinese University of Hong Kong
Support by central research funds from Hong Kong Hospital Authority and The Chinese University of Hong Kong. Gemcitabine provided by Eli Lilly.
E.P. Hui: Advisory Board: MSD, BMS; Speaker Honorarium: MSD. A.T.C. Chan: Research funding: Pfizer, Boehringer Ingelheim, BMS, MSD Travel, accomodation and expenses: Roche, Pfizer, Novartis, BMS. R.K.C. Ngan: Travel, accomodation and expenses: Eisai, Merck, Pfizer, Roche, Novartis. V.H. Lee: Honorarium: Roche, Eli Lilly, Pfizer. B.C. Zee: Patent, royalties: Health View Bioanalytics. K.C.A. Chan: Patents, royalties, other intellectual property: Sequenom, Illumina, Xcelom, Cirina. Stock or other ownership: Cirina, Xcelom. Honoraria, travel, accomodations, expenses: Biorad. Consulting or advisory role: Xcelom.
All other authors have declared no conflicts of interest.