The role of artificial intelligence in informed patient consent for radiotherapy treatments-a case report.

Recent advancements in large language models (LMM; e.g., ChatGPT (OpenAI, San Francisco, California, USA)) have seen widespread use in various fields, including healthcare. This case study reports on the first use of LMM in a pretreatment discussion and in obtaining informed consent for a radiation oncology treatment. Further, the reproducibility of the replies by ChatGPT 3.5 was analyzed. A breast cancer patient, following legal consultation, engaged in a conversation with ChatGPT 3.5 regarding her radiotherapy treatment. The patient posed questions about side effects, prevention, activities, medications, and late effects. While some answers contained inaccuracies, responses closely resembled doctors' replies. In a final evaluation discussion, the patient, however, stated that she preferred the presence of a physician and expressed concerns about the source of the provided information. The reproducibility was tested in ten iterations. Future guidelines for using such models in radiation oncology should be driven by medical professionals. While artificial intelligence (AI) supports essential tasks, human interaction remains crucial.

Impact of Surgical Margins in Breast Cancer After Preoperative Systemic Chemotherapy on Local Recurrence and Survival.

BACKGROUND: While "no tumour on ink" is an accepted margin width for R0 resection in primary surgery, it's unclear if it's oncologically safe after neoadjuvant chemotherapy (NAC). Only limited data demonstrate that surgery within new margins in cases of a pathological complete response (pCR) is safe. We therefore investigated the influence of different margins and pCR on local recurrence and survival rates after NAC. METHODS: We retrospectively analysed data of 406 women with invasive breast cancer, treated with NAC and breast-conserving therapy between 1994 and 2014 in two certified Austrian breast health centres. We compared R ≤ 1 mm, R > 1 mm and RX (pCR) for local recurrence-free survival (LRFS), disease-free survival (DFS) and overall survival (OS). RESULTS: After a median follow-up of 84.3 months, the 5-year LRFS (R ≤ 1 mm: 94.2%, R > 1 mm: 90.6%, RX: 95.0%; p = 0.940), the 5-year DFS (R ≤ 1 mm: 71.9%, R > 1 mm: 74.1%, RX: 87.2%; p = 0.245) and the 5-year OS (R ≤ 1 mm: 85.1%, R > 1 mm: 88.0%, RX: 96.4%; p = 0.236) did not differ significantly between narrow, wide, nor RX resections. Regarding DFS and OS, a negative nodal status reduced the hazard ratio significantly. CONCLUSION: There is no significant difference in LRFS, DFS and OS comparing close, wide or unknown margins after pCR. We suggest that resection in new margins after NAC is safe according to "no tumour on ink". Resection of the clipped area in cases of pCR is emphasized.

Comparison of PDR brachytherapy and external beam radiation therapy in the case of breast cancer.

Pulsed dose rate brachytherapy (PDR) was compared to external beam radiation therapy (EBRT) in the case of breast cancer. The benefits were figured out by evaluation of dosimetric parameters and calculating the normal tissue complication probability (NTCP). PDR plans were set up for five randomly chosen left-sided breast cancer patients delivering a total dose of 50.4 Gy to the target (dose rate 0.8 Gy h(-1)). For EBRT five left-sided breast cancer patients were planned using 3D-conformal tangential photon beams with a prescribed total dose of 50 Gy (2 Gy/fraction) to the total breast volume. For plan ranking and NTCP calculation the physical dose was first converted into the biologically effective dose (BED) and then into the normalized total dose (NTD) using the linear quadratic model with an alpha/beta ratio of 3 Gy. In PDR the relative effectiveness (RE) was calculated for each dose bin of the differential dose volume histogram to get the BED. NTCPs were calculated for the ipsilateral lung and the heart as contoured on CT slices based on the Lyman model and the Kutcher reduction scheme. Dosimetric parameters as V(th) (percentage of the total volume exceeding a threshold dose) and Jackson's f(dam) (fraction of the organ damaged) were also used to figure out the benefits. The comparison of calculated NTCPs in PDR and EBRT showed no difference between these two modalities. All values were below 0.01%. f(dam) derived from EBRT was always higher (mean value 8.95% versus 1.21% for the lung). The mean V(10) and V(20) of the lung related to BED were 6.32% and 1.72% for PDR versus 11.72% and 9.59% for EBRT. When using dosimetric parameters as V(th) and f(dam), PDR was mostly superior to EBRT in respect of sparing normal tissues. NTCP calculation as a single method of modality ranking showed a lack of information, especially when normal tissue was exposed to low radiation doses.