892P - Real-world data validation of university of San Diego nomogram for the prediction of benefit of intensive treatment for locoregionally advanced head and neck cancer

Background

Definitive concurrent chemoradiation (ChRT) +/- induction chemotherapy (ICT) is the only potentially curative strategy for unresectable locally advanced head and neck cancer (LAHNSCC) patients. However, it is an intensive treatment with high rate of acute and late toxicity. The nomogram developed in the University of San Diego (3 RTOG trials) and validated in the Veterans Affairs (VINCI dataset) allows risk stratification and identification of patients who benefit from intensive therapy (IT) (omega w score ≥ 0.80.) in overall survival (OS) through an online tool to calculate the w score. http://comogram.org//. Aim: to validate this nomogram in a Spanish population.

Methods

Clinical and treatment data from retrospective 290 LAHNSCC patients treated with definitive ChRT +/- ICT between 2010-2020 were collected. Performance of the nomogram for OS was assessed using: the area under the curve (AUC) and c-index (CI), Akaike information criterion (AIC), and the Nagelkerke test (of 0 to 1, the closer to 1 the better). Kaplan-Meier (log-rank test) and univariate Cox analysis to evaluate the interaction between IT (yes/no) and the w score (<0.8 or ≥0.8) were performed.

Results

170 patients obtained a w score ≥0.8. The application of the nomogram in our series obtained a moderate yield (AUC:0.68; CI:0.67; AIC:332, Nagelkerke:0.84). It was shown that an w score >= 0.8 is a statistically significant independent risk factor for OS (HR 2.058, p-value < 0.001). In the population with w score <0.8 patients treated with IT had not reached the median OS, while those without IT had an estimated median OS of 52.4 months. In contrast, in the population with w score ≥0.8, the median OS was 33.3 months (95% CI 2.7-66.3) in patients without IT, compared to 36.4 months (95% CI 22.8-49.3) in those patients treated with IT (p-value=0.005).

Conclusions

Our results support the utility of this nomogram in daily clinical practice in the Spanish population. It allowed an adequate stratification into 2 risk groups (high and low) based on the w score (<0.8 or ≥0.8) which was an independent prognostic factor, though we did not observe a differential effect of IT in these risk groups (benefit from IT in both).

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.