A three-drug induction beat the standard in nasopharyngeal cancer

For patients with high-risk, locally advanced nasopharyngeal carcinoma, replacing the long-standing two-drug induction regimen — cisplatin plus fluorouracil — with a three-drug combination of paclitaxel, cisplatin and capecitabine raised five-year failure-free survival from 62.9% to 77.6% [s1]. The gain, reported in NEJM Evidence after a median of more than seven years of follow-up, held across relapse, metastasis and death, but the more intensive regimen also caused more severe side effects [s1].

Nasopharyngeal carcinoma is a cancer of the upper throat that is strongly linked to Epstein–Barr virus and is far more common in parts of southern China and Southeast Asia than in Europe or North America. For disease that has spread to nearby lymph nodes or tissue but not distant organs, the standard approach is a short course of chemotherapy to shrink the tumour — "induction" — followed by chemoradiotherapy. The open question this trial set out to answer was not whether to give induction, but which drugs to use.

What the trial found

The trial randomly assigned 238 patients with high-risk locoregionally advanced disease to two 21-day cycles of either paclitaxel–cisplatin–capecitabine (TPC, 118 patients) or cisplatin–fluorouracil (PF, 120 patients), after which both groups received the same concurrent chemoradiotherapy [s1]. An earlier report from the same trial had already shown a three-year failure-free advantage for the three-drug arm [s2]; this analysis extends the picture to five years, with a median follow-up of 89.1 months [s1].

The primary endpoint, failure-free survival — time without relapse, metastasis or death — was 77.6% at five years in the TPC group versus 62.9% in the PF group, a hazard ratio of 0.52 (95% confidence interval, 0.34 to 0.82) [s1]. The secondary endpoints moved in the same direction. Five-year distant metastasis-free survival was 87.8% with TPC and 78.8% with PF (HR 0.51; 95% CI, 0.28 to 0.95); locoregional relapse-free survival was 92.0% versus 82.1% (HR 0.43; 95% CI, 0.21 to 0.88); and overall survival was 89.6% versus 82.2% (HR 0.51; 95% CI, 0.27 to 0.95) [s1]. That the benefit reached overall survival, not just the composite endpoint, is the result's strongest feature — survival is the outcome least vulnerable to how "failure" is defined.

The advantage was clearest in patients whose blood carried a lower burden of Epstein–Barr virus DNA before treatment. Among those with pretreatment EBV DNA below 3,000 copies per millilitre, five-year overall survival was 92.3% with TPC and 84.4% with PF; among patients with higher viral loads, the rates were closer, at 84.2% and 81.1% [s1]. Circulating EBV DNA behaves here much like the minimal residual disease that circulating tumour DNA is used to track in other cancers — a molecular readout of how much disease is present and how it is likely to behave.

The cost side of the ledger

Intensifying chemotherapy is rarely free. Grade 3 or 4 adverse events — the more serious categories — occurred in 68 patients, or 57.6%, of those given TPC [s1]. The trade-off in this trial fell on the favourable side, because the survival gains were large and durable, but a three-drug regimen is harder to tolerate than two, and that matters most for older or frailer patients who are underrepresented in trial populations.

How much to read into it

Several limits temper the finding. This is a single randomised trial of modest size, run at centres in a region where the disease is endemic and clinical experience with it is deep, so the results may not transfer cleanly to settings that treat few cases a year [s1]. The regimen swaps intravenous fluorouracil for oral capecitabine alongside adding paclitaxel, so the comparison bundles a new drug with a route change rather than isolating a single variable [s1]. And durable as the curves look, longer induction chemotherapy trials in this cancer have a history of the kind of drug docket where a strong early signal narrows with time — the reason the extended follow-up reported here matters as much as the original result [s2].

For now, the signal is consistent: in high-risk nasopharyngeal carcinoma, a paclitaxel-based triplet before chemoradiotherapy delivered a real and sustained survival advantage over the older doublet, at the price of more toxicity [s1]. Whether guidelines move to make it a standard, and how the choice is weighed against side effects for individual patients, is what to watch next. This article describes trial findings and is not medical advice.

Sources

• Induction Paclitaxel–Cisplatin–Capecitabine for Nasopharyngeal Carcinoma — NEJM Evidence, published online August 2026
• Effect of Induction Chemotherapy With Paclitaxel, Cisplatin, and Capecitabine vs Cisplatin and Fluorouracil in Nasopharyngeal Carcinoma — JAMA Oncology, 2022