Abstract: Objective Normofractionated (NF) radiation therapy (RT) is still standard for loco-regional early breast cancer (EBC) in many countries. HypoG-01, a UNICANCER, non-inferiority, open-label, multicenter, randomized phase III trial (NCT03127995), conducted in parallel with the DBCG Skagen trial 1 (NCT02384733), evaluated hypofractionated (HF) RT with 40 Gy/15 fr (2.67 Gy/fr) versus NF RT 50 Gy/25 fr (2.0 Gy/fr). This is the first report of oncological outcomes, secondary endpoints of the trial. Methods Patients (pts) ≥18 years old were operated for T1-3, N0-3, M0 breast cancer. All pts received nodal and thoracic wall or breast RT. Tumor-bed boost (sequential or simultaneous) and nodal levels treated were decided according to local guidelines. Target volumes were delineated according to the ESTRO consensus. RT technique was left at the investigator’s discretion. Stratification factors included treating center, type of surgery, number of positive nodes and body mass index. The primary endpoint, reported previously, was time to occurrence of arm lymphedema. Oncological outcomes included locoregional relapse free survival (LRFS), invasive disease-free survival (IDFS), distant disease-free survival (DDFS), breast cancer specific survival (BCSS) and overall survival (OS) as defined per DATECAN guidelines. All time to cancer related endpoints were defined as starting from the date of randomization until the event. The primary statistic test was stratified one-sided logrank test: 5% significance level in per-protocol population (PPP) with a pre-specified non-inferiority margin of 1.545. Stratified Cox model was used for calculating hazard ratios of oncological outcomes. Results In total, 1265 pts were randomized to HF versus NF RT from Sep 2016 to Mar 2020 with 1221 in the PPP (HF group 614 pts (50.3%); NF group 607 (49.7%)), 5 consent withdrawn and 39 major deviations. Median age was 58 years (range 23-91), surgery included mastectomy (501 pts; 45%) and axillary clearance (921 pts; 82.8%). Sequential (67.8%) or simultaneous integrated (32.2%) tumor-bed boost was used in 596 pts (48.8%). With a median follow-up for OS in PP analysis of 3.8 years (95% CI 3.3 to 3.9), HF was non-inferior to NF RT in terms of lymphedema (HR=1.07; 90% CI 0.86-1.34, non-inferiority p=0.003). Table 1 reports the hazard ratio and 3-year survival rates for the oncological endpoints LRFS, IDFS, DDFS, BCCS and OS. There was no sign of disadvantage for any of these endpoints comparing HF to NF. Additional analyses will be presented at the meeting. Conclusion Moderately HF loco-regional RT is non-inferior to NF RT in terms of lymphedema risk in EBC and does not show a disadvantage in terms of LRFS, IDFS, DDFS, BCCS and OS with a median follow-up of 3.8 years. Together with the Skagen 1 trial, this study provides level 1A evidence supporting the use of 40 Gy/15 fr for loco-regional radiation therapy in EBC with respect to arm lymphedema risk; Table 1: Oncological outcomes according to treatment arm in the per protocol analysis HF: Moderately hypofractionated radiation therapy; NF: Normofractionated radiation therapy; n:number of patients; N: number of events Citation Format: Sofia Rivera, Eleni Karamouza, Youlia Kirova, Séverine Racadot, Mohamed Benchalal, Jean-Baptiste Clavier, Claire Charra-Brunaud, Marie-Eve Chand-Fouche, Delphine Argo-Leignel, Karine Peignaux, Ahmed Benyoucef, David Pasquier, Philippe Guilbert, Julien Blanchecotte, Agnès Tallet, Adeline Petit, Guillemette Bernadou, Xavier Zasadny, Claire Lemanski, Jacques Fourquet, Emmanuelle Malaurie, Honorine Kouto, Carole Massabeau, Alexandre Henni, Pauline Regnault, Aurélie Belliere, Yazid Belkacemi, Magali Le Blanc-Onfroy, Julien Geffrelot, Jean-Briac Prevost, Marie Bergeaud, Assia Lamrani-Ghaouti, Amandine Ruffier, Naima Bonnet, Stephanie Wong, Christine Rossier, Thomas Brion, Pierre Maroun, Claire Petit, Guillaume Auzac, Stefan Michiels. First oncological outcomes of HypoG-01: a UNICANCER phase III trial comparing loco-regional hypo vs normo fractionated radiation therapy in early breast cancer patients [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-19-10.

Abstract: Introduction Cosmetic evaluation after breast cancer treatment is a clinical indicator of toxicity. User bias and inter-subject variability hamper this objective. To address this limitation, a deep learning approach was developed on the basis of the HYPOG-01 trial (NCT03127995), a phase III trial comparing hypo to normo-fractionated radiotherapy (RT) in breast cancer patients requiring nodal irradiation. Material and Methods Cosmetic outcomes using the Harris scale were assessed by a radiation oncologist using a 4-level rating system from excellent to poor. This evaluation involved photographs from 581 female patients included in the intention-to-treat population of the HYPOG-01 study analysis (mastectomy/pamectomy and non-usable cases excluded). Front images were taken with the arms along the body at baseline, 3-weeks after radiotherapy start, end of treatment, then 6 months and every year after randomization up to 5 years. Comparing manual landmark annotation with the semi-automated software BCCT.core©, the agreement rate was moderate, with an intra-class correlation coefficient (ICC) of 0.66 (95%CI 0.57-0.73). The dataset consisted of 2,348 images, which were divided into exclusive patient-based training (1,661), validation (308), and testing (377) datasets. The distribution of Harris scores in the dataset was highly imbalanced: 7% excellent, 33% good, 45% fair, and 15% poor. Nipple landmarks were used as landmarks to address picture acquisition variations by cropping and resizing images to 224 × 224 resolution. Feature extraction was performed using a Swin-TransformerV2, an advanced attention-based vision model initially trained on ImageNet. Newly integrated fully connected layers categorized the extracted features. The model was trained for 300 epochs, and the highest F1-score model was selected. Asymmetry in texture, marks, and breast geometry played a crucial role in Harris scoring. To improve the model's performance, we generated symmetrical images from the region of interest, averaging them with the original images, and we incorporated the timestamps from image captures as an additional influencing factor. We employed techniques such as contrast modulation, lighting adjustments, and geometric transformations to augment the dataset, introducing additional variations and enhancing the model's generalization and accuracy. Results The performance of our model was evaluated using balanced binary classification, multi-class accuracy, and F1-score. Comparatively, our model performed similarly to BCCT in terms of overall accuracy but demonstrated better performance in separating multiple classes, as indicated in Table 1. In Table 2, we present a confusion matrix that provides insights into the model's performance Conclusion The proposed solution simplifies and accelerates the evaluation process by utilizing only two nipple landmarks, surpassing manual and semi-automated tools. This advancement opens doors for automated, large-scale cosmetic toxicity evaluation. Continuous improvement and validation contribute to its robustness and reinforce its significant impact in assessing cosmetic outcomes after breast cancer treatment. Table. Evaluation of performance on the test set (only on cases with evaluation by BCCT (327 images)). Table. Confusion matrix between our predictions and the labels on the test set (only on cases with evaluation by BCCT (327 images)). Citation Format: Alexandre Cafaro, Amandine Ruffier, Gabriele Bielinyte, Youlia Kirova, Séverine Racadot, Mohamed Benchalal, Jean-Baptiste Clavier, Claire Charra-Brunaud, Marie-Eve Chand-Fouche, Delphine Argo-Leignel, Karine Peignaux, Ahmed Benyoucef, David Pasquier, Philippe Guilbert, Julien Blanchecotte, Agnès Tallet, Adeline Petit, Guillemette Bernadou, Xavier Zasadny, Claire Lemanski, Jacques Fourquet, Emmanuelle Malaurie, Honorine Kouto, Carole Massabeau, Alexandre Henni, Pauline Regnault, Aurélie Belliere, Yazid Belkacemi, Magali Le Blanc-Onfroy, Julien Geffrelot, Jean-Briac Prevost, Eleni Karamouza, Stefan Michiels, Marie Bergeaud, Assia Lamrani-Ghaouti, Sami Romdhani, Alexis Bombezin-Domino, Nikos Paragios, Sofia Rivera. Cosmetic assessment in the UNICANCER HypoG-01 trial: a deep learning approach [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-21-03.

Abstract: Use of circulating tumor DNA (ctDNA) for diagnosis is limited regarding the low number of target molecules in early-stage tumors. Human papillomavirus (HPV)–associated carcinomas represent a privileged model using circulating viral DNA (ctHPV DNA) as a tumor marker. However, the plurality of HPV genotypes represents a challenge. The next-generation sequencing (NGS)-based CaptHPV approach is able to characterize any HPV DNA sequence. To assess the ability of this method to establish the diagnosis of HPV-associated cancer via a blood sample, we analyzed ctHPV DNA in HPV-positive or HPV-negative carcinomas. Experimental Design: Patients (135) from France and Senegal with carcinoma developed in the uterine cervix (74), oropharynx (25), oral cavity (19), anus (12), and vulva (5) were prospectively registered. Matched tumor tissue and blood samples (10 mL) were taken before treatment and independently analyzed using the CaptHPV method. Results: HPV prevalence in tumors was 60.0% (81/135; 15 different genotypes). Viral analysis of plasmas compared with tumors was available for 134 patients. In the group of 80 patients with HPV-positive tumors, 77 were also positive in plasma (sensitivity 95.0%); in the group of 54 patients with HPV-negative tumors, one was positive in plasma (specificity 98.1%). In most cases, the complete HPV pattern observed in tumors could be established from the analysis of ctHPV DNA. Conclusions: In patients with carcinoma associated with any HPV genotype, a complete viral genome characterization can be obtained via the analysis of a standard blood sample. This should favor the development of noninvasive diagnostic tests providing the identification of personalized tumor markers. See related commentary by Rostami et al., p. 5158

Abstract: Purpose: In this paper, we describe the methodology used, and provide a first characterization of the study population and radiotherapy (RT) data in CANTO-RT (CANcer TOxicities Radiotherapy), the largest available multicenter prospective cohort of early breast cancer (BC) patients treated with RT that aims to identify predictors of development, and persistence of long-term toxicities. Methods: CANTO (NCT01993498) is a French prospective clinical cohort study of 10 150 patients with stage I-III BC from 26 cancer centers. Patients matching all CANTO inclusion and exclusion criteria, who received RT and were still in follow up, in the 10 top recruiting CANTO centers, with a minimum follow up of 3 years, were selected for CANTO-RT. Eligible patients had breast/chest wall +/- lymph node RT with curative intent. Individual full DICOM RT files (CT, RT Structure, RT Dose, RT Plan) were collected, anonymized, structured and analyzed on the CANTO-RT/UNITRAD web platform using AQUILAB Share Place™ and Analytics Dose module. Characteristics of the patients and tumors (including TNM, histology, HER2, estrogen and progesterone receptor) were recorded at baseline. Characteristics of the treatments, skin, lung, cardiovascular, neurological, musculoskeletal toxicities (CTCAE v4.0), QOL (BR23, QLQC30), cosmetic, and oncological outcomes were assessed at diagnosis (baseline), 3-6 (M0), 12 (M12), 36 (M36) and 60 (M60) months after completion of primary surgery, chemotherapy or radiotherapy whichever came last together, with blood, plasma and serum tests. Results: CANTO-RT enrolled 3875 BC patients between June 2012 and February 2017 with a median follow-up of 64 months :1947 (50.2%) left side, 1850 (47.8%) right side and 78 (2%) bilateral BC. The vast majority of patients had hormone receptor-positive tumors 3321 (85.7%) and 553 (14.3%) had human epidermal growth factor 2 (HER2) positive tumors; 2586 (66.7%) had stage pT1 and 2525 (65.2%) pN0 disease; 2087 (53.8%) neoadjuvant or adjuvant chemotherapy, 477 (12.3%) adjuvant trastuzumab and 3138 (81%) adjuvant endocrine therapy. Among 3797 patients with unilateral RT, 3065 (80.4%) had breast conserving surgery, 747 (19.6%) total mastectomy; 2712 (71.5%) sentinel node and 1080 (28.5%) axillary dissection. Tumor bed boost was delivered in 2658 patients (68.5%) and lymph node RT in 1356 patients (35%) including internal mammary chain in 844 patients (21.8%). Most patients 3691 (95.3%) were treated with 3D conformal RT and 184 (4.7%) with intensity-modulated RT. Normofractionated RT (2Gy/fraction) was mostly used (69.9%). Clinical target (breast, chest wall, lymph nodes) and contoured organs at risk (heart, left anterior descending coronary, lung, spinal cord, esophagus, thyroid, brachial plexus, contralateral breast, humeral head) contours and dose/volume histograms were automatically extracted after quality control procedure excluding corrupted files and inconsistencies 36 (1%) (Table 1). Conclusion: CANTO-RT is the largest early breast cancer prospective cohort with full individual clinical and DICOM RT data available. CANTO-RT is a valuable resource, open for collaborative projects, for identification and validation of clinical and dosimetric predictive factors of RT related toxicities. Further long term follow up is ongoing. Table 1.Baseline characteristics of the CANTO RT breast cancer patients.CharacteristicsBreast Cancer Patients [N(%) or Mean (range)]Age at enrolmentMean (range), years56.5 (23.3-85.8)Tumour size (pT)T037 (1)T12586 (66.7)T21058 (27.3)T3177 (4.6)Missing17 (0.4)Nodal status (pN)02525 (65.2)11035 (26.7)2223 (5.8)379 (2)Missing13 (0.3)Tumour histologyInfiltrating Ductal3011 (77.7)Lobular473 (12.2)Others (including mixed)381 (9.8)Missing10 (0.3)Hormone Receptors positiveNegative541 (14)Positive3321 (85.7)Missing13 (0.3)HER2Negative3305 (85.3)Positive553 (14.3)Missing17 (0.4)Type of chemotherapyNo chemotherapy1788 (46.1)Neoadjuvant chemotherapy450 (11.6)Adjuvant chemotherapy1629 (42)Peri-adjuvant chemotherapy (neo + adjuvant)8 (0.2)Hormonal therapyNo730 (18.8)Yes3138 (81)Missing7 (0.2)Herceptin treatmentNo or Not applicable3378 (87.2)Yes477 (12.3)Missing20 (0.5)Type of breast surgerylumpectomy3113 (80.3)Mastectomy734 (18.9)Right lumpectomy and Left mastectomy13 (0.3)Right mastectomy and Left lumpectomy9 (0.2)None6 (0.2)Type of lymph node surgerySentinel node2746 (70.9)Axillary dissection1086 (28)Right sentinel node, Left axillary dissection20 (0.5)Right axillary dissection, left sentinel node12 (0.3)None11 (0.3)Radiation therapyRight Side1850 (47.8)Left Side1947 (50.2)Bilateral78 (2.0)Patients with boostNo or Not applicable1217 (31.4)Yes2658 (68.6)Lymph node levels treatedNone2519 (65)Yes1356 (35)Level 1284 (20.9)Level 2340 (25.1)Level 31072 (79.1)Level 41348 (99.4)Internal mammary chain844 (62.2)Irradiation techniques3D3691 (95.3)IMRT184 (4.7)Fractionation regimensNormofractionation 25-fractions2707 (69.9)Hypofractionation 15-16 fractions166 (4.3)Hypofractionation and Partial breast irradiation51 (1.3)Unspecified fractionation - CTV breast or chest wall not delineated951 (24.5) Citation Format: Thomas Sarrade, Rodrigue Allodji, Youssef Ghannam, Guillaume Auzac, Sibille Everhard, Ophélie Querel, Youlia Kirova, Karine Peignaux, Philippe Guilbert, Claire Charra-Brunaud, Julien Blanchecotte, Rezart Belshi, David Pasquier, Séverine Racadot, Céline Bourgier, Sandrine Ducornet, David Gibon, Fabrice André, Florent De Vathaire, Sofia Rivera. CANTO RT: The largest prospective multicenter cohort of early breast cancer patients treated with radiotherapy including full DICOM RT data [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-19-01.

Abstract: Purpose: The aim of current analyses is to report toxicity and cosmetic outcomes at 3 and up to 9 years of follow-up of post-menopausal patients randomized to receive either standard external beam whole breast radiotherapy (WBI), including hypofractionated options, versus accelerated partial breast irradiation (APBI). Methods and materials: From December 2010 to July 2015, 1006 patients were enrolled in 34 French centers (503 in each arm). Among the whole population, 28 patients who did not meet the final selection criteria or withdrew consent were excluded leading to a modified intention to treat analysis dataset of 978 patients (WBI: n=488; APBI: n=490). Median age (65y) and tumor stage pT1 (99%) rates were similar in both arms. Patients had conservative surgery with clip placement in the tumor bed. Clear margins ( & gt; 2mm) were observed in 99% of the patients. In both arms, 96-97% of the patients had negative sentinel lymph node biopsy (SLNB; median number: 4 in WBI arm and 5 in APBI), luminal BC. Ductal histology was observed 82%. Only 2% and 1% of patients had grade III and pN(i+) disease. The median time interval between surgery and radiotherapy was 57d in WBI vs 62d in APBI. WBI schedules consisted of: 50Gy in 25fr + 16Gy boost (n=212) or 40Gy in 15fr (n=156) or 42.5Gy in 16fr (n=120), while APBI arm consisted of 38.5Gy or 40Gy in 10fr. Overall, 94 patients from the APBI arm finally received standard WBI. For statistical considerations, SHARE trial, sponsored by UNICANCER (NCT01247233) is a non-inferiority randomized controlled trial comparing APBI versus WBI in terms of local control as primary objective. Secondary endpoints were severe toxicity (NCI-CTCAE v4 grade ≥ 2), and cosmetic results, evaluated by doctors and by patients, over the entire follow-up. For both outcomes, we estimated the cumulative incidences (CI) using Kalbfleish and Prentice method, considering disease relapse, secondary cancer or death as competing events. Treatment effect (APBI vs WBI) was estimated by cause-specific Hazard Ratios (cs-HR) from Cox models adjusted on stratification factors. Results: Median follow-up was 5.8y (range, 0.13-9.5). The number of deaths was 27, and the number of local relapses was 8. Among the 978 patients, 582 and 396 had finally WBI and APBI, respectively. The rates of post-operative hematoma, edema and infection were low: 8-9%, 2%, 3-2%, respectively. When considering any type of severe toxicity, we observed a significant reduction rate in APBI compared to WBI: cs-HR=0.73 (95% confidence interval: 0.61-0.88); p=0.001, and 3-year cumulative incidence (CI) of severe toxicity at 45% (41-49) in WBI vs 36% (32-40) in APBI arm. The difference was also in favor of APBI when considering breast skin toxicity alone: cs-HR=0.55 (0.44-0.70), p & lt; 0.001 and 3-year CI at 36% (32-40) in WBI vs 21% (18-25) in APBI arm. Conversely, for breast other toxicities, WBI was found less toxic than APBI: cs-HR= 2.10 (1.51-2.91), p & lt; 0.001, and 3-year CI at 8% (5-10) vs 15% (12-19), respectively. When considering cosmetic results according to the investigator, we observed no significant difference between the two arms: cs-HR=1.04 (0.81-1.33), p=0.26 and 3-year probability of remaining with good to excellent cosmetic results at 77% (73-81) in WBI arm and 78% (74-81) in APBI arm. Findings were similar when considering results according to the patient: cs-HR=1.07 (0.85-1.37), p=0.23, and 3-year probability at 74% (70-78) and 75% (70-79), respectively. Conclusions Historically SHARE is the first APBI trial that included hypofractionated schedules in the standard arm. We reported increased risk of severe toxicity and skin breast toxicity in standard arm as compared with APBI arm without any difference in terms of cosmetic results. Longer follow-up is needed. Citation Format: Yazid Belkacemi, Isabelle Gabelle-Flandin, Marie-Cécile Le Deley, Adeline Petit, Philippe Guilbert, Julien Geffrelot, Christian Carrie, Eleonor Rivin Del Campo, Chantal Hanzen, Claire Charra-Brunaud, Isabelle Lecouillard, Nicolas Magne, Agnès Tallet, Nicolas Leduc, Blaha Belgadi, Philippe Fourneret, Alexandre Coutte, Esther Capelo, Franck Darloy, Muriel Garcia Ramirez, Philippe Dudouet, Pierre Clavere, Jean-Philippe Suchaud, Guillaume Auzac, Thomas Lacornerie, Jérôme Lemonnier, Céline Bourgier, Eric Lartigau. Early results of the French multicenter, randomized SHARE trial comparing whole breast irradiation versus accelerated partial breast irradiation in postmenopausal women with early-stage breast cancer. [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD3-05.