Bolus Versus Weekly Chemotherapy in Definitive Chemoradiation for Nasopharyngeal Cancer

Abstract

Ongoing randomized controlled trials for nasopharyngeal carcinoma (NPC) utilize administration of weekly concurrent cisplatin (cis), although there is limited data comparing its outcomes with bolus dosing. We therefore set out to compare efficacy of weekly versus bolus cisplatin in chemoradiation therapy (CRT) for the definitive treatment of NPC. We conducted a single institution retrospective review of patients undergoing definitive CRT for NPC with IMRT and/or protons between January 1, 2003, and December 31, 2016. Patients receiving induction chemotherapy or who did not receive cis were excluded. Dosing regimens were recorded as well as age, smoking status (≥10 pack-years vs <10), Charlson Comorbidity Index (CCI) (0, 1, or ≥2), stage grouping (I, II-III, IVA, or IVB), EBV (EBER) status and receipt of adjuvant chemotherapy. Bolus and weekly cis were grouped as intent-to-treat. The Kaplan-Meier method was used to compare locoregional control (LRC), progression free survival (PFS), distant metastasis (DM) and overall survival (OS). Univariate and multivariate Cox regression analysis were also performed to assess for factors impacting the same aforementioned outcomes. 159 patients were identified of whom 89 received bolus and 51 received weekly cis. Median age was 51.71% of patients were male. Median follow up was 53.5 months (range 2.7-147.0), 53.5 months for bolus and 16.9 months for weekly cis. 17 patients cumulatively received <200 mg/m2 of cisplatin, 8% of the bolus group and 17% of the weekly group. 98% of patients’ radiation therapy was delivered to a dose of at least 6996 cGy (or equivalent). 10 patients receiving bolus and 9 patients receiving weekly cis had documented locoregional failures. Two-year LRC was 95% with bolus and 79.2% with weekly cis (P =.005). Two-year PFS was 85.1% with bolus versus 69.2% with weekly cis (P = .064). DM or OS were not significantly associated with use of bolus versus weekly cis (P = .83 and .417 respectively). On multivariate analysis, both lower stage grouping (P = .003) and bolus cis (P = .008) were associated with significantly improved LRC, but not receipt of ≥200 mg/m2 cis (P = .297), age, smoking status, KPS, EBV or use of adjuvant chemo. There was improved PFS with lower stage grouping (P = .009), and lower CCI (.039) and a trend toward improvement with bolus cis (P = .069), but not receipt of ≥200 mg/m2 cis (P = .263). This single institution retrospective analysis demonstrated a significant improvement in 2-year LRC associated with bolus CRT compared to weekly that was independent to total dose within the narrow range of total dose administered within this cohort. The results are hypothesis-generating, and further prospective investigation into comparing the two regimens is likely warranted.