Back

Effectiveness of Radiation Therapy for Low- to Intermediate-Grade Neuroendocrine Tumors

Koenig, J. L.; Carter, J. N.; Aggarwal, S.; Chang, D. T.; Loo, B. W.; Diehn, M.; Koong, A. C.; Kunz, P. L.; Gensheimer, M. F.

2024-06-12 oncology
10.1101/2024.06.11.24308718 medRxiv
Show abstract

Surgery is the primary treatment for localized neuroendocrine tumors (NETs). Grade 1-2 NETs traditionally have been considered radioresistant due to their indolent nature but data regarding a role for radiation therapy (RT) are limited to old, small retrospective studies. We performed a retrospective review of patients with grade 1-2 NETs treated with RT at a large academic center to assess response and local failure rates. Radiographic response was evaluated with logistic regression. Local failure was assessed with cumulative incidence rates and competing risk regressions. We identified 35 patients with 78 treated lesions between 1974 and 2017. Most tumors originated from the pancreas (n=13) and bronchus/lung (n=11). Nine (26%) patients had grade 1 tumors, 16 (46%) had grade 2 tumors, and ten (29%) had grade 1-2 tumors. The median biologically effective dose (BED10) was 50.7 Gy (range, 20.0-106.5). The median follow-up was 13.5 months (range, 0.5 to 140.6 months). 20 of 21 (95%) patients had palliation of symptoms. Of 52 intact lesions, response was complete in 7 (13%), partial in 14 (27%), stable in 25 (48%), and progressive in 6 (12%) lesions. Higher BED10 was associated with a response (odds ratio/Gy 1.06; 95% CI, 1.02-1.11; p=0.008). Of 59 intact or resected lesions, the 2-year cumulative incidence of local failure was 26.4%. Grade 2 lesions were associated with local failure (hazard ratio 7.70; 95% CI, 1.22-48.8; p=0.03). We show that grade 1-2 NETs often respond radiographically and symptomatically to RT. RT should be considered in the management of grade 1-2 NETs.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.