Artificial intelligence to predict outcomes of head and neck radiotherapy

Head and neck radiotherapy induces important toxicity, and its efficacy and tolerance vary widely across patients. Advancements in radiotherapy delivery techniques, along with the increased quality and frequency of image guidance, offer a unique opportunity to individualize radiotherapy based on imaging biomarkers, with the aim of improving radiation efficacy while reducing its toxicity. Various artificial intelligence models integrating clinical data and radiomics have shown encouraging results for toxicity and cancer control outcomes prediction in head and neck cancer radiotherapy. Clinical implementation of these models could lead to individualized risk-based therapeutic decision making, but the reliability of the current studies is limited. Understanding, validating and expanding these models to larger multi-institutional data sets and testing them in the context of clinical trials is needed to ensure safe clinical implementation. This review summarizes the current state of the art of machine learning models for prediction of head and neck cancer radiotherapy outcomes.

Selective prophylactic lateral node dissection improves the ipsilateral lateral node recurrence-free survival: A retrospective single-center cohort study

BACKGROUND

Some studies have shown that prophylactic lateral neck dissection (pLND) may be beneficial for patients with papillary thyroid carcinoma (PTC); however, none of the Western guidelines currently recommends this procedure. Since 2007, the decision to perform pLND at our institution has been made on a case-by-case basis with different risk factors in mind. In this study, we investigated the significance and indications of pLND in patients with PTC.

METHODS

We identified patients at stage N0 or N1a and M0 with PTC who underwent surgery from 2007 to 2017. We compared lateral compartment recurrence-free survival (RFS) and distant RFS between patients who did and did not undergo pLND (pLND and non-pLND groups).

RESULTS

pLND was performed in 494/3177 (15.5%) patients with PTC (tumor size [T] ≥2 cm, cN0 or N1a, M0). Overall, no significant difference in lateral compartment RFS was detected between the pLND and non-pLND groups. On multivariate analysis, T ≥ 3 cm and positive extrathyroidal extension were independent predictors for recurrence to the lateral compartment. In the subset analysis of T ≥ 3 cm with positive extrathyroidal extension (n = 127), the lateral compartment RFS rate of the pLND group was significantly better (p < 0.01) than that of the non-pLND group (p < 0.01). In this subset, pLND reduced recurrence to the lateral compartment by 20.7% during the 5-year follow-up. However, pLND did not improve distant RFS.

CONCLUSIONS

pLND significantly improved lateral node RFS in patients having PTC ≥3 cm with significant extrathyroidal extension. For such patients, pLND at initial surgery may be considered to avoid second surgery.