ESTRO 2024 - Abstract Book

S1179

Clinical - Head & neck

ESTRO 2024

A total of 261 patients from 28 institutions of the JCOG Head and Neck Cancer Study Group were enrolled between 2012 and 2018. The JCOG1008 was terminated early for efficacy at the planned interim analysis due to predesigned advantage in experimental arm, so present results were analyzed earlier than originally planned. Of the 251 patients underwent radiotherapy, 87 and 164 were treated with 3DCRT and IMRT, respectively. Thyroid function tests were performed to assess hypothyroidism, including free thyroxine (FT4) and thyroid-stimulating hormone (TSH) assays. They were performed before enrolment and every 3 months during the first year after treatment, every 4 months during the second year, and every 6 months from the second to the fifth year. Hypothyroidism was defined as Grade 2 or higher in CTCAE v4.0. As candidates of predictive factors for hypothyroidism, sex (male or female), age, chemotherapy regimen (weekly or 3-weekly cisplatin), radiation delivery techniques (3DCRT or IMRT), primary tumor (oral, oropharynx, hypopharynx, larynx), presence or absence of positive lymph nodes and positive margins, baseline FT4, and TSH were extracted. For IMRT cases, we additionally extracted thyroid volume and the following thyroid dosimetric characteristics for thyroid: volume irradiated over 10 Gy (V 10Gy ), V 20Gy , V 30Gy , V 35Gy , V 40Gy , V 45Gy , V 50Gy , V 60Gy , mean dose, and maximum dose. We conducted the multivariable analysis on variables to examine predictive factors of hypothyroidism. A total of 162 patients (3DCRT; 57 patients and IMRT;105 patients) were included for the analysis, excluding protocol non-compliance (n = 10), death within 30 days from the last treatment date (n = 1), total thyroidectomy (n = 42), and partial thyroidectomy (n = 46). In addition, 97 IMRT patients excluding 8 patients without dosimetric data were included to compare dosimetric characteristics for thyroid. The median follow-up time was 4.7 years (range, 0.3-9.3 years). Of the 162 patients, 27 (16.7%) developed hypothyroidism within 2 years after the radiotherapy. Of the 3DCRT and IMRT patients, 7 (12.3%) and 20 (19.0%) developed hypothyroidism, respectively. The number at risk for 3DCRT and IMRT cases after radiotherapy was 57 (100.0%) and 102 (97.1%) at 3 months, 46 (80.7%) and 87 (82.9%) at 1 year and 38 (66.7%) and 65 (61.9%) at 2 years, respectively. There were no large differences between the two groups; however, the number at risk after 2 years was approximately 60% in the two groups, and its value was low. In a multivariable analysis for entire group, low baseline FT4 [OR = 0.051 (CI: 0.005-0.506, p = 0.011)] was found to be significantly associated with the development of hypothyroidism. Radiation delivery techniques showed no apparent association with the development of hypothyroidism [OR = 1.467 (CI: 0.543-3.961, p = 0.450)]. Female [OR = 2.497 (CI: 0.881-7.087, p = 0.085)] and the weekly cisplatin group [OR = 2.421 (CI: 0.951-6.165, p = 0.064)] tended to be associated with the development of hypothyroidism. For the 3DCRT group, there were no factors associated with hypothyroidism, while being female [OR = 5.543 (CI: 0.846-36.334, p = 0.074)] tended to be associated with the development of hypothyroidism. For the IMRT group, weekly cisplatin group [OR = 7.700 (CI: 1.632-36.343, p = 0.010)] and low baseline FT4 [OR = 0.009 (CI: <0.001-0.313, p = 0.010)] were found to be significantly associated with hypothyroidism. Of dosimetric characteristics, V 60Gy [OR = 1.069 (CI: 0.999-1.143, p = 0.054)] tended to be associated with the development of hypothyroidism. Results:

Conclusion:

To our best knowledge it is the first report of hypothyroidism in a prospective study of postoperative chemoradiotherapy for head and neck cancer patients with high-risk features. For entire group, baseline low FT4 was associated with adverse factors of hypothyroidism. Weekly cisplatin group and baseline low FT4 were associated with unfavorable factors of hypothyroidism among the IMRT group. We hope that future analysis at predesigned follow up at 3 years will provide more useful information.