Aim Artificial intelligence (AI) based auto-segmentation aids radiation therapy (RT) workflows and is being adopted in clinical environments facilitated by the increased availability of commercial solutions for organs at risk (OARs). In addition, open-source imaging datasets support training for new auto-segmentation algorithms. Here, we studied if the female and male anatomies are equally represented among these solutions. Materials and Methods Inquiries were sent to eight vendors regarding their clinically available OAR auto-segmentation solutions for each gender. The Cancer Imaging Archive (TCIA) was also screened for publicly available imaging datasets specific to the female and the male anatomy. Results All vendors provided AI based auto-segmentation solutions for the male pelvis and female breasts, while 5/8 vendors provided solutions for the female pelvis. The female breast and the female pelvis solutions were released at a median of 0.6 years and 2.3 years, respectively, after the release of the male pelvis solutions. Among 27 TCIA datasets identified, 15 involved the female anatomy (breast: 10; pelvis: 5) and 12 involved the male pelvis but no female-specific dataset included OAR segmentations, while three male pelvis datasets included OARs (ejaculatory duct, neurovascular bundle, penile bulb and verumontanum). Conclusion Commercial AI auto-segmentation solutions and open-source imaging datasets include considerably more solutions and OAR segmentations for male cancer over female cancer sites. This gender disparity is likely to propagate throughout the RT pipeline.

Purpose/Objective(s) Meta analysis of randomized trials in early-stage, node-negative breast cancer patients, showed an absolute mortality reduction of 3% with the addition of adjuvant radiation (RT). In early-stage endometrial cancer (EMCA), available data from randomized trials don't demonstrate such a reduction. Arguably the true impact of adjuvant RT, may have been obscured by including patients with low-risk features, lack of consistent surgical staging, and the prevalence of comorbidities in patients with early-stage EMCA enrolled in these randomized trials. The purpose of this analysis is to evaluate the impact of adjuvant RT on surgically staged, FIGO 2009 stage IB grade 3 and stage II endometrioid EMCA. Materials/Methods Retrospective study of 349,404 patients with EMCA in the US between 2004 and 2012 derived from the National Cancer Database (NCDB). Strict inclusion criteria included patients with surgically staged, FIGO 2009 stage IB grade 3 or stage II endometrioid carcinoma who underwent a total hysterectomy (TH) with a Charlson-Deyo-Comorbidity-Index of 0. Patients whose surgery wasn't TH, those prior cancer history, and those who received adjuvant chemotherapy were excluded. Univariable (UV) and multivariable (MV) logistic regression models for odds of receiving RT were conducted. Propensity-score matched pairs from the RT and no RT cohorts were identified using a 1:1 nearest neighbor matching algorithm with a caliper width of 0.005, matching on age, grade, stage, and LVI. UV and MV Cox Proportional Hazards Models were conducted for time to death (OS) in the matched cohort, stratified on the matched pairs. Kaplan Meier analysis was used to estimate 5-year OS. Results 3,286 patients met the inclusion criteria, 959 (29%) did not receive RT, and 2,327 (71%) patients received RT. Of the 2,327 who received RT, 957 patients were included in the propensity score matched analysis. Using the stratified log-rank test, the 5-year mortality rate was 21% (95% CI = 18-24) in the RT group compared to 24% (CI: 21-27), p = 0.001. MV Cox Model showed that the risk of death is 1.49 times higher for patients who did not receive RT compared to those who received RT ([1.14, 1.95], p value = 0.004). Patients with a salary of < $68,000 have a hazard of death 1.82 times higher compared to patients with a salary >$68,000 ([1.17, 2.84], p value = 0.008). Conclusion This study found an absolute mortality reduction of 3% with the addition of adjuvant RT in surgically staged FIGO 2009 stage IB grade 3 and II endometrioid EMCA, like that in early-stage breast cancer. The other independent predictor of higher mortality rate was income < $68,000. Such findings may help future investigators in designing treatment interventions/trials.

Purpose/Objective(s) Deep inspiration breath hold (DIBH) is widely used for radiotherapy of abdominal cancers to minimize the motion of the target and organs at risk during imaging and treatment delivery. During treatment, however, many patients demonstrate difficulties in maintaining the breath-hold level that they achieved voluntarily at simulation. This study investigates patient specific factors and characteristics in the breathing trace at simulation that can predict intolerance to DIBH during treatment, thus providing insights for fine tuning motion management strategies to each patient’s capacity. Materials/Methods We retrospectively studied 86 patients who received radiotherapy between January and July 2022 for abdominal cancers at a single institution. DIBH was implemented using a real time infrared monitoring system that tracks the motion of a reflective marker on the anterior thoracic surface. Patients received coaching by a radiation therapist to achieve a reproducible DIBH within 3 mm anteroposterior (AP) motion. A reference trace was recorded at simulation showing the marker displacement during DIBH and the free-breathing portions before and after DIBH. The study endpoint was any changes relative to the reference trace that were necessary for treatment. This was detected from the DIBH traces acquired during treatment. Correlation tests were conducted between the endpoint and 4 metrics extracted from the trace at simulation (AP levels at DIBH and during the expiration phase of the free breathing portion of the trace, and the period and amplitude of the free breathing portion). Additionally, 3 clinical and demographic factors at baseline were analyzed (age, Karnofsky Performance Scale, patient preferred language). Results Thirty percent (26/86) of the patients required modification to the planned DIBH protocol, including 22 DIBH level change for at least 1 fraction and 4 re-simulations under free-breathing. There was no statistically significant association at 5% false discovery rate (FDR) between the 7 investigated factors and DIBH protocol modification. Marginally significant associations were found for the DIBH AP level at simulation (FDR = 0.06, odds ratio = 1.06/mm, 95% confidence interval: 1.01/mm – 1.12/mm) and non-English language as a preferred language (FDR = 0.15, odds ratio = 4.52, 95% confidence interval: 0.99 – 20.61). Conclusion Despite the lack of statistical significance, the patient specific factors that are most predictive of DIBH tolerance were DIBH AP level and patient non-English language preference, which was in line with our clinical experience. To enhance patient tolerance to DIBH, patients need to be coached to avoid overly deep inspiration. Moreover, language barrier can have a potential impact, which may be mitigated by visual coaching.

UH-RT+/- IO for melanoma of the LGT is well-tolerated with local disease control achieved in 61% of patients. Vulvar primary and c-Kit mutations were associated with worse local control. Additional follow-up may be necessary to determine the value of dose escalation.

11005 Background: Despite progress in the enrollment of underrepresented in medicine (UIM) individuals in medical schools, hematology/oncology (HO) remains one of the least diverse medical subspecialties. We evaluated current recruitment efforts and barriers to the inclusion of UIM individuals in HO fellowship programs. Methods: We conducted a national online survey of program directors (PDs) and associate PDs (reported in aggregate as PDs) of adult and pediatric HO fellowships. Each PD received a unique survey link via email and 4 email reminders over the 8-week study period. Interim results at 4-weeks are presented and summarized with descriptive statistics. Results: At a pre-defined interim time point of 4 weeks, 78 HO PDs participated (18% response rate) and 69 completed the survey (88% completion rate). Most self-identified as women (57%, n = 38) and White (55%, n = 37), were affiliated with an academic program (79%, n = 53), and led an adult HO fellowship (73%, n = 49). Most PDs reported under or no representation of UIM faculty in their divisions; 9% (n = 6) self-identified as UIM. PDs reported their HO fellowships matched a median of 3 UIM fellows over the last 5 years (IQR 2-5 over 5 years); 46% (n = 32) reported UIM recruitment was unchanged and 4% (n = 3) decreased during this period. About half of PDs (51%, n = 35), matched a UIM candidate in the 2023 match. Of the 21% (n = 14) of PDs whose programs had no UIM fellows, 0% matched a UIM applicant in the 2023 match. 52% (n = 36) of PDs endorsed an inability to match UIM applicants despite consistently ranking them highly. PDs perceive the top 3 barriers to UIM recruitment as: low number of UIM applicants (53%, n = 41), geographic program location (50%, n = 39), and lack of diverse faculty/fellows in their program (42%, n = 31). 10% (n = 7) also reported meeting resistance about DEI initiatives. While 53% (n = 41) of PDs believe fellow demographics should reflect the local population, 23% (n = 16) don’t know how to increase diversity and 11% (n = 8) do not believe their program’s diversity needs to increase. In answers to open ended questions, some PDs expressed disagreement with the definition of UIM and reported a perceived limited number of qualified UIM applicants. Almost all institutions provide DEI training (96%, n = 72) for faculty and most PDs believe it has improved recruitment and inclusion of diverse individuals (52% and 65%, respectively). Yet, 34% (n = 23) agree that UIM fellows in their program are treated differently because of their race/ethnicity and 22% (n = 15) agree that UIM fellows repress their identity or background to be respected by faculty, peers, and staff. Conclusions: Historically and currently excluded populations remain underrepresented in the HO workforce. PDs face multiple barriers to the recruitment of diverse fellows, including differing views on the definition of UIM and perception of the value of diversity in the physician workforce.

11003 Background: Research has shown workplace sexual harassment (WSH) is a pervasive problem in academia, affecting recruitment, retention & career advancement of women faculty. This study examines WSH in academic oncology faculty in a national survey. Methods: We obtained data from the Association of American Medical Colleges StandPoint Faculty Engagement Survey between 2020-22 representing a national sample of academic faculty from 25 medical schools (overall response rate (RR) 58%). WSH was considered specific “unwanted behavior” by coworkers that occurred at least once in the past year. Reporting questions were asked on a 5-point Likert Scale from Strongly Disagree to Strongly Agree, with 6th option “I Don’t Know.” Responses were grouped as Agree/Strongly Agree vs all others. Chi-square tested for differences by gender and race/ethnicity. Results: 807/1,287 oncologists participated in the survey (RR 63%), and 91% (731) answered the WSH questions. Most self-identified as men (58%) and only 8% identified as underrepresented race/ethnicity (URiM). See Table for findings. In oncology, women experienced any WSH at higher rates than men (33.1% vs 10.3%, p=0.001). The most common was derogatory comments due to gender. Regardless of gender, most knew how to report WSH incidents, but men were more likely to feel safe reporting. While 64.9% of men felt confident that their medical school would resolve WSH incident(s) effectively, only 48.7% of women did. Positive perceptions of reporting decrease among those who reported harassment. URiM faculty (35.6% vs.18.5%, p=0.001) were more likely than non-URiM faculty to experience at least one of the five harassment behaviors, although this was limited by small sample. Conclusions: In academic oncology, women and URiM experience workplace sexual harassment more often than men and non-URiM. Less than half of women felt confident that reporting an incident would lead to effective resolution. Urgent intervention is needed to combat the negative force of sexual harassment on workforce equity. [Table: see text]

Purpose/Objective(s) To assess physics quality assurance (PQA) and practices in less resourced RT clinics to improve the quality of care. Materials/Methods An initial survey was conducted in 2020 to thirteen select RT centers in six countries. In 2021, from the survey outcome, our team conducted onsite visits to all the RT-centers in one of the countries: one private, and two public (Public-1- and –2). Public-1 in the capital and Public-2 in a regional capital. Follow-up surveys were sent to 17 medical physicists. Questions centered on the topic of equipment, institutional practice, physics quality assurance, management, and safety practices. Qualitative and descriptive methods were used for data analysis. We accessed operational challenges using a 5-point Likert system: machine downtime, patient related issues, power outage and staffing. Results We received a 100% response rate. Six (6/17) had 1-3 years of experience and two (2/17) have over 15 years of experience. Over 80% of respondents had experience with a treatment planning system. Private and public-1 have CT simulators located in RT clinic. Breast, genitourinary, Gynecologic, H&N were the most treated diseases in all three centers. All three (3) clinics have 3D conformal RT (3D-CRT), the private and public-2 have electron treatment modalities and only public-1 have IMRT treatment available. An audit during the visit showed that no single point diode patient specific QA is done for 3D-CRT patients and none of the institutions had a functioning diode to carry out this QA process. The average daily patient workload on external beam device for each clinic was LINAC: Private = 25, public-1 = 55, public-2 = 40; Co- 60: public-1 = 45, Public-2 = 25. Public-1 and-2 lacked the equipment necessary to conform to best practices in TG-142 and 198. Public-2 reported significant challenges in the operational challenges above with values of 4.5, 3 and 2.75 and 2.25. Notably Public-1 and-2 have peer review chart rounds which are attended by clinical oncologists, medical physicists, physicians, and physics trainees. All (17/17) responded to having a system of documenting, tracking, and trending patient related safety incidents but only one (1/17) physicist reported using a voluntary incident reporting system. Although not quantifiable audits during site visits show education and training remains the most important need in operating successful local PQA and management programs. In the initial survey RT centers in lower GNIpc country there was a direct correlation between QA needs and the country index. Conclusion The initial study showed a direct correlation between QA needs and the development index of a country which has led to the first of a continent-wide survey intended to spotlight PQA practices in LIMCs, the challenges faced, lessons learned, to help understand the gaps and needs to better support their local PQA and management programs.

Our survey results show that most (75%) RO PDs believe that UIM persons do not have equal access to tertiary education in the US (17% disagreed and 8% were neutral), but that UIM residents are supported in their department (78%). Most PDs reported that they do not participate in specific activities aimed at recruiting UIM residents, with nearly 1 in 5 programs reporting no interviews of UIM applicants in the last 5 years, and over one quarter of programs acknowledging differential treatment based on race/ethnicity. These data suggest a potential disconnect between DEI perceptions and activities among RO PDs that may be addressed through increased awareness and tailored programs.

C. Weil: None. F. Lew: None. D.K. Gaffney: Research Grant; NCI. Consultant; NCI. run meetings; NCI.G. Suneja: Research Grant; National Institutes of Health. Honoraria; National Comprehensive Cancer Network. Travel Expenses; National Comprehensive Cancer Network, Radiation Oncology Institute, American Board of Radiology; Radiation Oncology Institute, National Comprehensive Cancer Network, ASTRO.V.M. Williams: Involved in planning the mission and activities for the subcommittee; ARRO Equity and Inclusion Subcommittee.

Purpose/Objective(s)Workforce studies demonstrate that representation of Black, Latinx, Indigenous people and women in radiation oncology (RO) is lacking and relates to structural inequities, bias, and the systemic exclusion of diverse individuals in science, technology, engineering, and mathematics (STEM). To that end, ASPP was launched in 2019 with the goal of exposing STEM undergraduate and early medical students who are women and/or students from racial and ethnic backgrounds that are historically underrepresented in RO to the field and its related specialties, including physics and radiation biology. In October 2020, this 3-hour program was held virtually during ASTRO's Annual Meeting. Here we describe the results of a pre- and post-survey administered to ASPP registrants.Materials/MethodsUndergraduates and medical students were recruited broadly from known STEM programs with focus on UIM (underrepresented in medicine) and women students. Registrants were emailed pre- and post-program surveys via Qualtrics. Surveys collected demographic information regarding their perception of RO graded on a 5-point Likert scale (1-very negative to 5-very positive). Questions were asked on topics previously found to influence medical students’ choice of specialization (e.g., work hours, diversity, patient interaction, impact on patient outcomes, job security). There were also questions unique to the post-survey to help assess the impact of ASPP on attendees’ consideration of RO as a specialty.ResultsOne hundred ninety-four students registered for ASPP and 95 registrants completed the pre-survey (49% response rate). 47/95 (49%) of respondents were undergraduates and 51% were medical students. 72% identified as female and 28% male. The racial distribution was 35% Black/African American, 31% White, 19% Asian, 2% American Indian, and 0% Native Hawaiian/Pacific Islander. 20% of respondents identified as Hispanic. For geographic breakdown of registrants: 27% Midwest, 19% Northeast, 51% South and 2% West. 62% of respondents had no prior dedicated RO exposure, and 78% responded that they knew “a little” or “not much” about what radiation oncologists do. 31/65 attendees completed the post-survey (48% response rate). When comparing pre- and post-surveys regarding their perceptions of predetermined topics, mean responses increased favorably for diversity in RO 3.0 vs 3.7, work/life balance 4.0 vs 4.6, and patient interaction 4.2 vs 4.8, respectively. 100% of post-survey respondents said they would seek a RO rotation/elective in their remaining medical training and 93% would consider applying for the ASTRO Minority Summer Fellowship.ConclusionThe ASPP virtual program reached a diverse group of underrepresented STEM students and positively influenced their perceptions of and interest in RO. Based on the success of ASPP, we hope to continue this program annually. Future targeted efforts should be made to increase UIM interest in RO.

Functional imaging (DWI) has shown promise in predicting response and tumor control in cervical cancer treated with radiation/chemotherapy, but clinical translation has been limited by complexities in image registration, tumor delineation on apparent diffusion coefficient (ADC) maps, and artifacts. Two methods, scanner-generated auto-coregistration (Coaut) and more resource-intense individualized coregistration (Coind) may influence DWI results. Our purpose was to compare both methods' consistency and accuracy for tumor response prediction and assess potential pitfalls. We analyzed 21 FIGO stage IB2-IVA cervical cancer patients enrolled in a prospective imaging trial who were treated with concurrent chemoradiation. Coaut vs. Coind were studied in each with 3 serial MRIs, pre-therapy (pre), early-therapy (2-wk) and mid-therapy (5-wk) obtained. ADC data sets contained T2- weighted (T2w)-based whole-tumor contouring and were linked to 1-month post-therapy tumor response. For Coind, the scanner-DICOM-generated Coaut were adjusted individually to account for distortion and shifts of the tumor region between T2w and DWI. Intensity histogram tumor features, mean, coefficient of variation, skewness, and kurtosis were extracted from the Coaut and Coind MRI ADC images at each time point and correlated with treatment response. Pairwise consistency of features was evaluated with signed rank testing. Predicted response classification was compared by ROC area-under-the-curve (AUC) and rank sum testing. Pre and 2-wk mean ADC were consistent (<20% variability, p>0.34) between Coaut and Coind in most patients (20/21 each); 5-wk mean ADC was less consistent (18/21). Inconsistencies were due to tumor distortion/shifts (4 patients) and artifact (1). Coind-based pre and 5-wk mean ADC predicted response (AUC 0.81 and 0.82, p = 0.021 and 0.017, respectively); Coaut-based pre mean ADC showed similar performance (AUC = 0.81, p = 0.021) but not 5-wk mean ADC (AUC 0.65, p = NS). Coind-based decrease in 2-wk ADC skewness best predicted response (AUC = 0.84, p = 0.013), outperforming Coaut (AUC 0.71, p = NS). Inconsistencies between scanner automated and individualized co-registration were small in the vast majority of cases for whole-tumor based pre-therapy mean ADC, which predicted response with both methods. Major ADC distortion and ADC- T2w misalignments that were easily identifiable benefited from additional individualized co-registration adjustment. Pre-therapy DWI MRI for response prediction in cervical cancer is readily exportable to community settings using scanner based post-processing. This capability may broaden the application of DWI for treatment of cervical cancer in the wider community. Supported by R01CA155454.

Advances in radiotherapy technique have improved tumor control and reduced toxicity in the management of nasopharyngeal carcinoma (NPC). Local failure remains a problem for some patients with very advanced primary tumors, and both acute and late toxicities are significant given the large anatomic volume treated and close proximity of tumor to numerous critical structures. Proton therapy (PT) offers dosimetric advantage over IMRT, which can further improve the therapeutic ratio. There is still limited clinical data reported. We report our early clinical outcomes with PT for locally advanced NPC. We reviewed treatment records of patients enrolled on a prospective IRB-approved clinical registry study who received PT for definitive treatment of NPC. Demographics, dosimetry, disease control outcomes, and acute and late toxicities were reviewed (CTCAE v.4). Analyses were done using descriptive statistics. 21 patients were treated from 2015-2018. There were 6 females and 15 males, with median age 57 (range 19-73). T stage distribution was T1, 4; T2, 4; T3, 1; T4, 12. N stage distribution was N0, 2; N1, 7; N2, 10; N3, 2. 95% had stage III-IV disease. WHO classification was type 1 in 6, type 2/3 in 14, unknown in 1. 71% were EBV positive. A dose-painted pencil beam scanning approach was used with 2-5 beams encompassing primary and bilateral neck. Majority was treated to 69.96 CGE, in 33 fractions once daily; 2 underwent hyperfractionated accelerated treatment twice daily. All received concurrent CDDP delivered either every 3 weeks, 17 or low dose weekly, 4; 5 also received induction chemotherapy. With median follow up of 16 mos (range 1-42 and 14 patients with >1 year), 18 patients are NED, 1 developed local failure who had a WHO I, T4 primary with intracranial extension and 2 developed distant metastases. Locoregional control rate was 95%, DM free rate 90%, and OS was 90%. Acute toxicities included grade 3 mucositis in 14, grade 3 dermatitis in 9. PEG utilization included prophylactic, 10; reactive, 4; none, 7. Late toxicities included grade 2 xerostomia in 2, hearing loss in 3, and only 1 patient remains PEG-dependent. Proton therapy is feasible in locally advanced NPC with early outcomes demonstrating excellent locoregional control and favorable toxicity profile. This is consistent with early results from other single institutions. Longer follow up and additional comparative studies are needed to evaluate the relative advantages compared to IMRT.