Machine learning-based automated planning for hippocampal avoidance prophylactic cranial irradiation

PURPOSE

Design and evaluate a knowledge-based model using commercially available artificial intelligence tools for automated treatment planning to efficiently generate clinically acceptable hippocampal avoidance prophylactic cranial irradiation (HA-PCI) plans in patients with small-cell lung cancer.

MATERIALS AND METHODS

Data from 44 patients with different grades of head flexion (range 45°) were used as the training datasets. A Rapid Plan knowledge-based planning (KB) routine was applied for a prescription of 25 Gy in 10 fractions using two volumetric modulated arc therapy (VMAT) arcs. The 9 plans used to validate the initial model were added to generate a second version of the RP model (Hippo-MARv2). Automated plans (AP) were compared with manual plans (MP) according to the dose-volume objectives of the PREMER trial. Optimization time and model quality were assessed using 10 patients who were not included in the first 44 datasets.

RESULTS

A 55% reduction in average optimization time was observed for AP compared to MP. (15 vs 33 min; p = 0.001).Statistically significant differences in favor of AP were found for D98% (22.6 vs 20.9 Gy), Homogeneity Index (17.6 vs 23.0) and Hippocampus D mean (11.0 vs 11.7 Gy). The AP met the proposed objectives without significant deviations, while in the case of the MP, significant deviations from the proposed target values were found in 2 cases.

CONCLUSION

The KB model allows automated planning for HA-PCI. Automation of radiotherapy planning improves efficiency, safety, and quality and could facilitate access to new techniques.

Intraoperative irradiation in breast cancer: preliminary results in 80 patients as partial breast irradiation or anticipated boost prior to hypo-fractionated whole breast irradiation

Purpose

To present the first results of intraoperative irradiation (IORT) in breast cancer with a low-energy photon system used as partial breast irradiation (PBI) or as an anticipated boost before whole breast hypo-fractionated irradiation (IORT + WBI), concerning tolerance, side effects, quality of life, and patient-reported outcomes.

Materials and methods

Eighty patients treated with an Intrabeam^® system of 50 kV X-rays received a 20 Gy dose intraoperatively were included. Moderate daily hypofractionation of 2.7 Gy in 15 fractions up to 40.5 Gy was administered if high-risk factors were present. Acute post-operative toxicity, surgery complications, chronic toxicity, patient-reported cosmesis and Breast-Q questionnaire were performed at follow-up visits.

Results

Thirty-one patients were treated as PBI and the remaining 49 as IORT + WBI. Only the IORT + WBI group presented acute toxicity, mainly mild acute dermatitis (11 patients) and one subacute mastitis. A total of 20 patients presented fibrosis (18 patients grade I, 2 patients grade II), 15 (30.5%) patients in the IORT + WBI group and 3 (9.6%) patients in the group of PBI. The cosmesis evaluation in 73 patients resulted poor, fair, good or excellent in 2, 7, 38 and 26 patients, respectively. In PBI group Breast-Q scored higher, especially in terms of their psychosocial well-being (78 vs 65) and satisfaction with radiation-induced toxicity (77 vs 72, respectively) compared to IORT + WBI group.

Conclusion

IORT is a well-tolerated procedure with low toxicity, good cosmesis and favorable patient-reported outcomes mainly when administered as PBI.

Predictive factors for survival in neoadjuvant radiochemotherapy for advanced rectal cancer

INTRODUCTION

Neoadjuvant radiochemotherapy followed by radical surgery is the standard approach in advanced rectal carcinoma. Tumor response is determined in histological specimen.

OBJECTIVE

To assess predictive factors for survival in 115 patients.

PATIENTS AND METHOD

115 patients treated with neoadjuvant radiochemotherapy followed by radical surgery with total mesorectal excision, in our hospital from January 2007 to December 2014. All patients received pelvic radiotherapy with concomitant chemotherapy, followed by radical surgery and in some adjuvant chemotherapy.

RESULTS

In univariate analysis, distance to anal verge, radial margin, perineural invasion, and good grade regression are predictive factors for both, specific and disease free survival; and in multivariant, only radial margin and perineural invasion were predictive factors for survival. We found distance to anal verge (<5 cm) as the only clinical factor to predict a positive margin in the histologic specimen.

CONCLUSIONS

Perineural invasion and positive radial margin are predictive factors for both specific and disease free survival.

Predictors of acute esophagitis in lung cancer patients treated with concurrent three-dimensional conformal radiotherapy and chemotherapy

PURPOSE

To evaluate the risk factors for acute esophagitis (AET) in lung cancer patients treated with concurrent 3D-CRT and chemotherapy.

METHODS AND MATERIALS

Data from 100 patients treated with concurrent chemoradiotherapy with a mean dose of 62.05 +/- 4.64 Gy were prospectively evaluated. Esophageal toxicity was graded according to criteria of the Radiation Therapy Oncology Group. The following dosimetric parameters were analyzed: length and volume of esophagus in treatment field, percentage of esophagus volume treated to >or=10, >or=20, >or=30, >or=35, >or=40, >or=45, >or=50, >or=55, and >or=60 Gy, and the maximum (D(max)) and mean doses (D(mean)) delivered to the esophagus. Also, we developed an esophagitis index (EI) to account the esophagitis grades over treatment time.

RESULTS

A total of 59 patients developed AET (Grade 1, 26 patients; Grade 2, 29 patients; and Grade 3, 4 patients). V50 was associated with AET duration (p = 0.017), AET Grade 1 duration (p = 0.016), maximum analgesia (p = 0.019), esophagitis index score (p = 0.024), and AET Grade >or=1 (p = 0.058). If V50 is <30% there is a 47.3% risk of AET Grade >or=1, which increases to 73.3% if V50 is >or=30% (p = 0.008). The predictive abilities of models (sensitivity and specificity) were calculated by receiver operating characeristic curves.

CONCLUSIONS

According to the receiver operating characeristic curve analysis, the 30% of esophageal volume receiving >or=50 Gy was the most statistically significant factor associated with AET Grade >or=1 and maximum analgesia (A(max)). There was an association with AET Grade >or=2 but it did not achieve statistical significance (p = 0.076).

Prevention of radiochemotherapy-induced esophagitis with glutamine: results of a pilot study

PURPOSE

To assess the usefulness of oral glutamine to prevent radiochemotherapy-induced esophagitis in patients with lung cancer, and to determine the dosimetric parameter predictive of esophagitis.

METHODS AND MATERIALS

Seventy-five patients were enrolled; 34.7% received sequential radiochemotherapy, and 65.3% received concomitant radiochemotherapy. Every patient received prophylactic glutamine powder in doses of 10 g/8 h. Prescribed radiation doses were 45-50 Gy to planning target volume (PTV)1 (gross tumor volume plus wide margins) and 65-70 Gy to PTV2 (reduced margins). The primary endpoint was the incidence of Grade 2 or greater acute esophagitis.

RESULTS

No patient experienced glutamine intolerance or glutamine-related toxicity. Seventy-three percent of patients who received sequential chemotherapy and 49% of those who received concomitant chemotherapy did not present any form of esophagitis. V50 was the dosimetric parameter with better correlation between esophagitis and its duration. A V50 of <or=30% had a 22% risk of esophagitis Grade >or=2, which increased to 71% with a V50 of >30% (p = 0.0009).

CONCLUSIONS

The use of oral glutamine may have an important role in the prevention of esophageal complications of concomitant radiochemotherapy in lung cancer patients. However, randomized trials are needed to corroborate that effect. V50 is the dosimetric parameter with better correlation with esophagitis grade and duration.