A systematic review and proportional meta-analysis of image-based pattern of loco-regional failure analyses outcomes in head and neck squamous cell carcinoma.

BACKGROUND AND PURPOSE

The prognosis following loco-regional failure after primary radiotherapy (RT) for head and neck squamous cell carcinoma (HNSCC) is poor. The hypothesis that most failures occur as a consequence of tumor radioresistance, can be evaluated by proxy as the proportion of failures that occur in the high-dose region. Several studies have investigated possible reasons for treatment failure by an image-based pattern of failure analyses (POF), comparing the initial planning CT scan with a scan conducted upon failure. The aim of the present systematic review and meta-analysis was to evaluate the proportion of failures that occurred in the high-dose region of all analyzed failures.

MATERIALS AND METHODS

A systematic database search from 2000 to 2023, was performed for studies including results from image-based loco-regional POF, regardless of the method, after primary RT for HNSCC. Proportions of volumetrically in-field (opposed to marginal or outfield) failures, point of origin-based inside high-dose targets, or covered by curative doses for both the number of patients and the number of failure sites were analyzed in proportional meta-analyses. The review was registered at Prospero (CRD42023412545).

RESULTS

Out of 56 included studies, accumulated image-based POF results were available from 1,161 patients and 658 individual failure sites. The majority of patients had in-field failures in volumetric-based studies (84 % (95 % CI: 77;90)), inside failures in point of origin-based studies (82 % (95% CI:61;85)) or failures covered by 95 % of dose prescribed to CTV1 (84 % (95% CI:69;95)). A trend toward increasing proportions of non-high-dose failures in more recently treated patients was observed.

CONCLUSION

Most loco-regional failures for patients treated with primary RT for HNSCC are related to the high-dose volume. Therefore, a focus on biomarkers predicting individual tumor radiosensitivity is warranted to enable individualized treatment intensification to increase loco-regional control.