Background
Retrospective cohort analysis using the National Cancer Data Base (NCDB), a hospital-based registry representing ~70% of all US cancer cases. 2,210 patients with resected high-risk major salivary gland carcinomas (T3–T4, N1–N3, or positive margins; grade 2–3) treated 1998–2011. Compared overall survival outcomes for adjuvant concurrent chemoradiotherapy (CRT) vs radiotherapy (RT) alone after resection.
Interventions and follow up
Arm A: Resection + adjuvant concurrent chemoradiotherapy (chemotherapy within 14 days of RT start), N=368 (16.7%)
Arm B: Resection + adjuvant radiotherapy alone, N=1,842 (83.3%)
Primary endpoint: Overall survival
mFollow up: Median 39 month
Arm B: Resection + adjuvant radiotherapy alone, N=1,842 (83.3%)
Primary endpoint: Overall survival
mFollow up: Median 39 month
Results
Unadjusted 2-year OS: 71.3% (CRT) vs 80.2% (RT alone)
Unadjusted 5-year OS: 38.5% (CRT) vs 54.2% (RT alone); HR 1.51 (95% CI 1.29–1.76), P<.001
Multivariate analysis: CRT associated with inferior OS (HR 1.22, 95% CI 1.03–1.44, P=.02)
Propensity-matched analysis: CRT HR 1.20 (95% CI 0.98–1.47, P=.08) — not statistically significant
Unadjusted 5-year OS: 38.5% (CRT) vs 54.2% (RT alone); HR 1.51 (95% CI 1.29–1.76), P<.001
Multivariate analysis: CRT associated with inferior OS (HR 1.22, 95% CI 1.03–1.44, P=.02)
Propensity-matched analysis: CRT HR 1.20 (95% CI 0.98–1.47, P=.08) — not statistically significant
Adverse events
Main adverse events: Not formally assessable in NCDB database. Higher-risk patients selected for CRT likely had more treatment-related toxicity.
Conclusions
The addition of concurrent chemotherapy to adjuvant RT in resected high-risk major salivary gland carcinoma did not improve overall survival and was associated with potentially worse outcomes in unadjusted and multivariate analyses. No subgroup analysis (by age, T stage, N stage, margin status, or histology) demonstrated a benefit from adding chemotherapy.
Key Limitations
Key Limitations: Major selection bias: sicker patients with more aggressive disease were more likely to receive CRT, which likely confounds the survival comparison despite propensity matching. NCDB lacks data on RT doses, chemotherapy regimens, and radiation field design — significant heterogeneity likely. Cannot account for comorbidities and ECOG performance status. Histology-specific analysis limited by aggregated data. The finding should not be interpreted as CRT causing harm, but rather that selection bias makes the benefit impossible to demonstrate in a retrospective database. No prospective randomized trial exists for adjuvant CRT in salivary gland carcinoma.
Clinical Context
Despite this negative NCDB analysis, concurrent cisplatin-based CRT is frequently used for high-risk salivary gland carcinomas (positive margins, ECE, high-grade histology) by analogy with other H&N SCC data. The lack of prospective RCT evidence means management remains center-dependent. NCCN currently lists CRT as an option for high-risk features. A prospective trial (NRG HN005) has explored de-escalation in HPV+ oropharynx but not salivary gland — this remains an unmet need in salivary oncology.
References