Background:Circulating tumor DNA (ctDNA) dynamics have emerged as a promising biomarker of treatment response in oncology drug development. Early decreases in ctDNA levels after treatment initiation are associated with improved long-term outcomes in patients with advanced cancer. To be considered for regulatory decision-making, patient-level meta-analyses evaluating individual-(I-) and trial-(T-) associations between early ctDNA changes and clinical outcomes using randomized controlled trial (RCT) datasets are necessary. Methods:While not fit-for-purpose, the Friends of Cancer Research ctMoniTR Project patient-level dataset included 10 RCTs in advanced cancer that were aggregated. Cox proportional hazards models assessed I-associations between molecular response (MR) and overall survival (OS) or progression-free survival (PFS) using either 90% decrease (MR90) or clearance (MR100) as MR cutoffs. T-associations compared treatment effects on MR with treatment effects on OS or PFS. Results:In pooled analyses across all trials, MR90 and MR100 were significantly associated with improved OS (MR90 adjusted hazard ratio [aHR] = 0.51, 95% CI 0.44-0.59; MR100 aHR = 0.45, 95% CI 0.39-0.53; both p < 0.0001) and improved PFS (MR90 aHR = 0.62, 95% CI 0.54-0.71; MR100 aHR = 0.56, 95% CI 0.48-0.64; both p < 0.0001). Associations were generally consistent across cancer types and treatment modalities. Trial-level associations between MR and OS were weak (R2 ~ 0.08-0.13), whereas associations between MR and PFS were stronger, particularly in aNSCLC (R2 up to 0.74). Conclusions:Early decreases in ctDNA were consistently associated with improved clinical outcomes at the individual-level across advanced cancers. Although trial-level associations were modest, stronger relationships with PFS support continued prospective evaluation of ctDNA dynamics as potential early endpoints in oncology drug development. Additional work using datasets that prospectively included plasma collection for ctDNA analyses is warranted.

PURPOSE: The purpose of this study was to evaluate the use of telemedicine amid the SARS-CoV-2 pandemic in patients with cancer and assess barriers to its implementation. PATIENTS AND METHODS: Telehealth video visits, using the Houston Methodist MyChart platform, were offered to patients with cancer as an alternative to in-person visits. Reasons given by patients who declined to use video visits were documented, and demographic information was collected from all patients. Surveys were used to assess the levels of satisfaction of treating physicians and patients who agreed to video visits. RESULTS: Of 1,762 patients with cancer who were offered telehealth video visits, 1,477 (83.8%) participated. The patients who declined participation were older (67.7 v 60.2 years; P < .0001), lived in significantly lower-income areas (P = .0021), and were less likely to have commercial insurance (P < .0001) than patients who participated. Most participating patients (92.6%) were satisfied with telehealth video visits. A majority of physicians (65.2%) were also satisfied with its use, and 74% indicated that they would likely use telemedicine in the future. Primary concerns that physicians had in using this technology were inadequate patient interactions and acquisition of medical data, increased potential for missing significant clinical findings, decreased quality of care, and potential medical liability. CONCLUSION: Oncology/hematology patients and their physicians expressed high levels of satisfaction with the use of telehealth video visits. Despite recent advances in technology, there are still opportunities to improve the equal implementation of telemedicine for the medical care of vulnerable older, low-income, and underinsured patient populations.

15 Background: BESPOKE CRC, a multicenter, prospective, observational study investigated the clinical utility of ctDNA for detecting MRD-based early recurrence in pts with surgically resected CRC. The primary endpoint was to assess the impact of ctDNA testing on treatment decisions and asymptomatic recurrence rates. The secondary endpoint was to assess the MRD clearance rate, survival of MRD-negative pts, overall survival, and patient-reported outcomes. Methods: Complete clinical and laboratory data were available for 1001 pts with stages II–III CRC. Longitudinal ctDNA testing was performed prospectively using a clinically validated, personalized, tumor-informed 16-plex mPCR-NGS assay (Signatera, Natera, Inc.). Plasma time points (n=8,536) were collected during the MRD (2-12 weeks postoperatively) and surveillance (post-adjuvant chemotherapy [ACT] completion/12 weeks postoperatively for pts observed) windows. We evaluated the correlation between ctDNA status and disease-free survival (DFS) as part of exploratory analysis. Results: Following curative resection, 62.4% (625/1001) pts received ACT: 25.9% (115/443) stage II, and 91.3% (510/558) stage III.Among pts with ctDNA results available during the MRD window with a median follow-up of 23.15 (range: 2.89-33.45) months, ctDNA-positivity was observed in 8.1% (34/420) of stage II and 24.9% (126/505) of stage III pts. A significant association between ctDNA positivity and inferior DFS was observed in stage II (HR=10.4; p<0.0001), and stage III (HR=10.1, p<0.0001) pts. 18/24 month DFS estimates for stages II-III combined were: (MRD-negative: 93.0%/91.7%; MRD-positive: 44.4%/41.4%). Analysis of surveillance was performed separately for pts observed (N=368) vs. ACT-treated (N=597). During surveillance, in the observation cohort, 6.8% (22/323) of stage II, and 33.3% of stage III (15/45) pts tested ctDNA-positive correlating with significantly worse DFS (stage II: HR=34.9; p<0.0001, stage III: HR=34.1; p=0.0008). Likewise, in the ACT cohort, 10.9% (12/110) of stage II and 21.1% of stage III (103/487) pts turned ctDNA-positive and had significantly worse DFS (stage II: HR=131.4; p<0.0001, stage III: HR=54.6; p<0.0001). Analysis of primary and secondary endpoints will be presented. Conclusions: ctDNA positivity was highly prognostic of DFS within MRD and surveillance windows in a subset of stages II-III pts enrolled in the BESPOKE CRC study. Our results highlight the potential value of ctDNA-based MRD detection for treatment-decision making, and findings relevant to clinical utility will be reported in the conference presentation. Clinical trial information: NCT04264702 .

Background: This study evaluates the effects of hydroxytyrosol (HT), a component of olive oil, on mammographic breast density reduction. We explored effects of HT on Wnt β-catenin and other pathways involved in cancer stem cell renewal, DNA repair, cell proliferation, and differentiation. Methods: Twenty-five milligrams per day oral dose of HT was given for 12 months in pre- and postmenopausal women at increased risk of breast cancer. Out of 51 patients enrolled, 41 completed the study. The annualized percent decrease in maximum mammographic volumetric breast density (max VBD%) between baseline (BL) and end of treatment (EOT) was analyzed. RNA sequencing (RNA-Seq) and multiplex analysis was performed on the breast biopsies to compare the BL with EOT samples. Results: Max VBD% showed a nonsignificant change; however, in women 60 years or older, the max VBD% decrease was significant (3.7%, p = 0.0391), especially in those with high BL mammographic density. Using RNA-Seq, 3330 unique transcripts were identified (p < 0.05). Mitotic telophase/cytokinesis and DNA damage were upregulated, whereas Wnt, Notch, and oxidative stress-induced senescence pathways were downregulated (p < 0.05). These pathways were confirmed by NanoString nCounter where significant decrease in proliferative genes (RELA and CDK4) and Wnt pathway (R-HSA-195721 and R-HAS-201681) was observed (p < 0.05). Conclusions: HT reduced breast density only in women over 60 years, especially in those with high BL breast density. HT also reduced proliferation and affected the Wnt signaling pathway. This study lays the foundation for future larger studies in exploring a natural compound with well tolerability and overall nontoxic profile for chemoprevention of breast cancer. Trial Registration: ClinicalTrials.gov identifier: NCT02068092.

BACKGROUND:Circulating tumor DNA (ctDNA) is a promising intermediate end point for oncology drug development, potentially accelerating regulatory approvals by providing early insights into treatment response. However, challenges remain in standardizing ctDNA assessment, including optimal blood collection timing and defining molecular response (MR) cutoffs. The ctDNA for Monitoring Treatment Response (ctMoniTR) project, led by Friends of Cancer Research, aggregates patient-level data from clinical trials to evaluate associations between ctDNA changes and overall survival (OS). METHODS:This analysis included four randomized clinical trials of patients with advanced non-small cell lung cancer (aNSCLC) treated with either anti-programmed death (ligand) 1 (anti-PD(L)1) therapy (with or without chemotherapy) or chemotherapy alone. MR was assessed using three predefined per cent-change thresholds in ctDNA levels (≥50% decrease, ≥90% decrease, and 100% clearance). ctDNA samples were analyzed at two timepoints: an early window (T1, up to 7 weeks post-treatment initiation) and a later window (T2, 7-13 weeks post-treatment initiation). Multivariable Cox proportional hazards models and time-dependent analyses were used to evaluate associations between ctDNA changes and OS. RESULTS:A total of 918 patients were included. In the anti-PD(L)1 group, ctDNA reductions at both T1 and T2 were significantly associated with improved OS across all MR thresholds. In the chemotherapy group, associations were weaker at T1 but became more pronounced at T2. Patients with MR at both T1 and T2 had the strongest OS associations. Overall, the results suggest that T2 had marginally stronger association with OS than T1. CONCLUSIONS:This study supports the potential of ctDNA as an intermediate end point in aNSCLC, with MR at both early (T1) and later (T2) timepoints showing significant associations with OS. Differences in ctDNA dynamics between treatment modalities highlight the importance of considering the timing of blood collection. Further research is needed to determine the optimal time window for assessing ctDNA response. Prospective trials and trial-level meta-analyses will be critical to validating ctDNA as a regulatory-grade intermediate end point for oncology drug development.

9 Background: ctDNA-based post-surgical detection of molecular residual disease (MRD) is known to be predictive of a high risk of recurrence. Here, we report the first results of BESPOKE CRC, a multicenter, prospective, observational study evaluating the ability of a tumor-informed ctDNA assay to inform ACT treatment decisions in stage II/III CRC patients (pts). Methods: Of the 1792 pts enrolled between 2020-07-02 and 2022-08-25, plasma samples from the first 350 pts with stage II-III CRC were analyzed. ctDNA was detected and quantified using a personalized, tumor-informed assay (Signatera, Natera, Inc.). Following curative resection, 232 pts received ACT and 118 underwent observation. Results: The cohort included 154 stage II and 196 stage III CRC pts; the median follow-up was 24.8 months. ctDNA results at the post-op MRD time point (tp) were available for 295 pts; 15.6% (46/295; stage II: 9/130=6.9%; stage III: 37/165=22.4%) of pts were ctDNA positive (ctDNA+) at MRD tp (MRD+). MRD-positivity was significantly associated with inferior disease-free survival (DFS) in stages II-III combined (HR=20.8, 95% CI: 10.0-43.4, p<0.0001) and in stage-stratified subgroups (stage II: HR=25.7, 95% CI 6.8-96.7; stage III: HR=18.1, 95% CI 7.3-45.1). Within the MRD+ group, pts receiving ACT had longer DFS compared to those in the observation group (median DFS: 18.7 vs 6.7 months; HR=3.9, 95% CI: 1.3-11.5, p=0.01). In contrast, no benefit of ACT was observed in MRD- pts (HR=1.1, 95% CI: 0.3-3.9, p=0.89). Of the MRD+ pts, 39.1% (18/46) had ctDNA clearance at 12-weeks post-surgery tp. Pts with ctDNA clearance had longer DFS compared to those who remained positive (median DFS: 24.2 vs 13.8 months; HR=0.4, 95% CI 0.1-1.0, p=0.045), however, had worse DFS than pts who were ctDNA- at both 4- and 12-weeks (HR=22.5, 95% CI: 6.8-75.0, p<0.0001). Notably, 44.4% (8/18) pts with ctDNA clearance recurred; all 8 turned back ctDNA+ before radiological detection of relapse. ctDNA results during surveillance were available for 339 pts, of whom 8.3% (58/339) were ctDNA+ and had significantly worse DFS compared to serially ctDNA- pts (HR=124.3, 95% CI: 29.8-518.7, p<0.0001). Conclusions: ctDNA-based MRD detection of MRD was highly prognostic of recurrence in an early representative subset of BESPOKE CRC cohort. Data from the expanded cohort will be presented at the meeting. ctDNA MRD results were also predictive: significant benefit from ACT was observed in MRD+ but not in MRD- pts. Additionally, early ctDNA clearance in response to adjuvant therapy and ctDNA status during surveillance were prognostic of pt outcomes. Our results highlight the potential utility of ctDNA-guided adjuvant therapy in pts with stage II/III CRC. The results of BESPOKE CRC herein, as one of the first ctDNA-based prospective studies, will be further validated by ongoing ctDNA-directed randomized clinical trials. Clinical trial information: NCT04264702 .

Purpose To evaluate the performance of a biopsy decision support algorithmic model, the intelligent-augmented breast cancer risk calculator (iBRISK), on a multicenter patient dataset. Materials and Methods iBRISK was previously developed by applying deep learning to clinical risk factors and mammographic descriptors from 9700 patient records at the primary institution and validated using another 1078 patients. All patients were seen from March 2006 to December 2016. In this multicenter study, iBRISK was further assessed on an independent, retrospective dataset (January 2015–June 2019) from three major health care institutions in Texas, with Breast Imaging Reporting and Data System (BI-RADS) category 4 lesions. Data were dichotomized and trichotomized to measure precision in risk stratification and probability of malignancy (POM) estimation. iBRISK score was also evaluated as a continuous predictor of malignancy, and cost savings analysis was performed. Results The iBRISK model's accuracy was 89.5%, area under the receiver operating characteristic curve (AUC) was 0.93 (95% CI: 0.92, 0.95), sensitivity was 100%, and specificity was 81%. A total of 4209 women (median age, 56 years [IQR, 45–65 years]) were included in the multicenter dataset. Only two of 1228 patients (0.16%) in the "low" POM group had malignant lesions, while in the "high" POM group, the malignancy rate was 85.9%. iBRISK score as a continuous predictor of malignancy yielded an AUC of 0.97 (95% CI: 0.97, 0.98). Estimated potential cost savings were more than $420 million. Conclusion iBRISK demonstrated high sensitivity in the malignancy prediction of BI-RADS 4 lesions. iBRISK may safely obviate biopsies in up to 50% of patients in low or moderate POM groups and reduce biopsy-associated costs. Keywords: Mammography, Breast, Oncology, Biopsy/Needle Aspiration, Radiomics, Precision Mammography, AI-augmented Biopsy Decision Support Tool, Breast Cancer Risk Calculator, BI-RADS 4 Mammography Risk Stratification, Overbiopsy Reduction, Probability of Malignancy (POM) Assessment, Biopsy-based Positive Predictive Value (PPV3) Supplemental material is available for this article. Published under a CC BY 4.0 license. See also the commentary by McDonald and Conant in this issue.

Abstract Probability of malignancy (POM) for Breast Imaging Reporting and Data System (BI-RADS) category 4 designated breast lesions ranges from 2% – 95% and contributes to a high unnecessary biopsy rate. This is as most clinicians often stick to the biopsy option to rule in or out breast cancer early; withholding biopsy could be risky, and biopsies of BI-RADS 4 lesions serve as a quality metric and performance standard. At 21.1%, biopsy-proven positive predictive value (PPV3) rates for BI-RADS 4 have not improved for decades, translating to high false-positive rates of mammography. Unnecessary biopsies are a big issue in the management of BI-RADS 4 lesions with negative implications including increased medical costs, healthcare wastes, unnecessary psychological burdens to the patients, and potential complications and risks. Objectives 1. Optimize the precision breast cancer risk assessment tool, iBRISK, that utilizes artificial intelligence (AI) technologies, including natural language processing (NLP), image processing, and deep learning, with clinical risk factors and imaging features. 2. Develop a user-friendly web interface for iBRISK to facilitate clinicians or insurers in estimating cancer risk and making informed biopsy decisions for BI-RADS 4 lesions. Methods Our intelligent-augmented breast cancer risk calculator (iBRISK) model was trained on multimodal data collected from 10,778 patients, including demographic factors, historical and clinical characteristics, mammographic features, and pathologic signatures. We validated iBRISK using 4,200 patients from multiple leading hospitals, including Houston Methodist Neal Cancer Center, the University of Texas MD Anderson Cancer Center, and the University of Texas Health San Antonio MD Anderson Mays Cancer Center. The iBRISK was connected to a backend server which is linked to a frontend web user interface using technologies like Hypertext Preprocessor (PHP) for the backend application running on an APACHE HTTP server and React-JavaScript for the front-end web application that communicates with the backend using Representational State Transfer Application Programming Interface (RESTful API) and JavaScript Object Notation (JSON) format. The communication was secured using Secure Sockets Layer (SSL) encryption. Results The iBRISK model demonstrated high sensitivity in malignancy prediction and achieved an accuracy of 89.5%, area under the receiver operating characteristic curve of 0.93 (95% CI: 0.92-0.95), sensitivity of 100%, and specificity of 81%. Only 0.16% of lesions determined to have low POM by the model were malignant. Our multi-center study shows that iBRISK achieves at least 50% reduction in unnecessary biopsies of BI-RADS 4 cases. Data elements for the required 20 features are entered into the interface using a variety of imputation methods including direct text, dropdown menus and radio button selections. The user-friendly web interface provides risk scores (0 – 1), risk levels (low, medium, and high), and associated biopsy recommendations. Conclusion The user-friendly iBRISK web interface is proposed as an adjunct to the BI-RADS system, enhancing the precision of BI-RADS 4 lesion cancer risk stratification. It is expected to reduce unnecessary biopsies, lower health costs, and enhance the quality of health care. This approach aims to tackle a critical issue in breast cancer diagnosis by leveraging advanced AI technologies and big data and providing clinicians with a tool to make more informed and precise biopsy decisions for BI-RADS 4 lesions. Citation Format: Chika Ezeana, Xiaohui Yu, Zhihao Wan, Tiancheng He, Tejal Patel, Virginia Kaklamani, Maryam Elmi, Erika Brigmon, Pamela Otto, Kenneth Kist, Lin Wang, Joe Ensor, Heather Speck, Ya-Chen Shih, Bumyang Kim, I-Wen Pan, David Spak, Wei Yang, Jenny Chang, Stephen Wong. An on-line deep learning decision support tool, iBRISK, aimed at improving breast cancer risk estimation and reducing unnecessary biopsies for BI-RADS 4 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 PO2-28-06.

Background: The survival benefit of adjuvant chemotherapy after surgical resection of oligometastases from colorectal cancer (CRC) remains unclear. The prognostic role of circulating-tumor DNA (ctDNA) was reported recently and a risk stratification strategy based on monitoring minimal/molecular residual disease (MRD) has been proposed, however, which drug regimen is most effective for ctDNA-positive patients is unknown. Methods/Design: Oligometastatic CRC patients planning to undergo surgery were registered in this study. After metastasectomy, the registered patients were enrolled in the treatment arm, in which 8 courses of modified-FOLFOXIRI (mFOLFOXIRI; irinotecan 150 mg/m2, oxaliplatin 85 mg/m2, l-leucovorin (l-LV) 200 mg/m2, and 46-h continuous infusion of 5-fluorouracil (5-FU) 2400 mg/m2 every 2 weeks) followed by 4 courses of 5-FU/l-LV are administered. The patients who did not meet the eligibility criteria for the treatment arm or did not consent to mFOLFOXIRI enrolled in the observation arm in which standard of care treatment is provided. Prospective blood collections for retrospective ctDNA analysis are scheduled pre-surgery, and at 28 days, 4 and 7 months after surgery. The primary endpoint is treatment compliance at 8 courses of mFOLFOXIRI and the key secondary endpoints are the ctDNA-positivity rate and survival outcomes in ctDNA-positive and -negative groups. A total of 85 patients will be enrolled from 11 institutions. First patient-in was on July 2020. Accrual completed in February 2024. Discussion: This study will potentially identify a better treatment strategy for patients with resectable oligometastatic CRC having postsurgical ctDNA positivity, compared to the current standard of care approaches.

BackgroundWhether non-genetic prognostic factors significantly influence the variable prognosis of antipsychotic-induced weight gain (AIWG) has not yet been systematically explored. MethodsSearches for both randomized and non-randomized studies were undertaken using four electronic databases, two trial registers, and via supplemental searching methods. Unadjusted and adjusted estimates were extracted. Meta-analyses were undertaken using a random-effects generic inverse model. Risk of bias and quality assessments were undertaken using Quality in Prognosis Studies (QUIPS) and Grading of Recommendations Assessment, Development and Evaluation (GRADE), respectively. ResultsSeventy-two prognostic factors were assessed across 27 studies involving 4426 participants. Only age, baseline body mass index (BMI), and sex were suitable for meta-analysis. Age (b=-0.044, 95%CI -0.157-0.069), sex (b=0.236, 95%CI -0.086-0.558), and baseline BMI (b=-0.013 95%CI -0.225-0.200) were associated with nonsignificant effects on AIWG prognosis. The highest quality GRADE rating was moderate in support of age, trend of early BMI increase, antipsychotic treatment response, unemployment, and antipsychotic plasma concentration. Trend of early BMI increase was identified as the most clinically significant prognostic factor influencing long-term AIWG prognosis. ConclusionsThe strong prognostic information provided by BMI trend change within 12 weeks of antipsychotic initiation should be included within AIWG management guidance to highlight those at highest risk of worse long-term prognosis. Antipsychotic switching and resource-intensive lifestyle interventions should be targeted toward this cohort. Our results challenge previous research that several clinical variables significantly influence AIWG prognosis. We provide the first mapping and statistical synthesis of studies examining non-genetic prognostic factors of AIWG and highlight practice, policy, and research implications.

Supplementary Table from A Phase 2 Trial of Enhancing Immune Checkpoint Blockade by Stereotactic Radiation and In Situ Virus Gene Therapy in Metastatic Triple-Negative Breast Cancer

Supplementary Tables 1-5 from 14-3-3ζ Overexpression Defines High Risk for Breast Cancer Recurrence and Promotes Cancer Cell Survival

Introduction: BI-RADS category 4 is associated with a wide variability in probability of malignancy, ranging from 2 to 95% while biopsy-derived positive predictive value (PPV3) for this category’s lesions remains low at 21.1% in the US. A major fallout of these facts is that we have way very high false positive rate leading to too many unnecessary biopsies and their associated costs and emotional burden. We improved our in-house intelligent-augmented Breast cancer RISK calculator (iBRISK), an integrated deep learning (DL) based decision support app and assessed its performance in a multicenter IRB-approved study. Methods: We improved iBRISK by retraining the DL model with an expanded dataset of 9,700 patient records of clinical risk-factors and mammographic descriptors from Houston Methodist Hospital (HMH) and validated using another 1,078 patient records. These patients were all seen between March 2006 and December 2016. We assessed the model using blinded, independent retrospective BI-RADS 4 patients who had biopsies subsequently after mammography and seen January 2015 - June 2019 at three major healthcare institutions in Texas, USA: MD Anderson Cancer Center, the University of Texas Health Sciences Center at San Antonio, and HMH. We dichotomized and trichotomized the data to evaluate precision of risk stratification and probability of malignancy (POM) estimation translated into biopsy decision augmentation. The iBRISK score as a continuous predictor of malignancy and possible cost savings was also analyzed. Results: The multicenter validation dataset had 4,209 women, median age (interquartile) was 56 (45, 65) years. The use of iBRISK score as a continuous predictor of malignancy yielded an AUC of 0.97. Among “low” and “moderate” POM patients, only two out of 1,228 patients (0.16%) and 118 out of 1788 (6.6%) were malignant respectively. This translates to an even better precision when compared to newly introduced BI-RADS 4 subcategories 4A and 4B, with associated PPV3s of 7.6% and 22%, respectively. The “high” POM group had a malignancy rate of 85.9% (1,025/1,193). Estimated potential cost savings in the US was over $260 million. Conclusion: The iBRISK app demonstrated high sensitivity in malignancy prediction and can potentially be used to safely obviate biopsies in up to 50% of patients in low/moderate POM-groups. This would result in significant healthcare quality improvement, cost savings, and help reduce patient anxiety. Citation Format: Chika F. Ezeana, Tiancheng He, Tejal A. Patel, Virginia Kaklamani, Maryam Elmi, Erica Ibarra, Pamela M. Otto, Kenneth A. Kist, Heather Speck, Lin Wang, Joe Ensor, Ya-Chen T. Shih, Bumyang Kim, I-Wen Pan, David Spak, Wei T. Yang, Jenny C. Chang, Stephen T. Wong. A multicenter study validated an integrated deep learning model for precision malignancy risk assessment and reducing unnecessary biopsies in BI-RADS 4 cases. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5698.

Supplementary Methods and Materials from 14-3-3ζ Overexpression Defines High Risk for Breast Cancer Recurrence and Promotes Cancer Cell Survival

Abstract Introduction: BI-RADS category 4 is associated with a wide variability in probability of malignancy, ranging from 2 to 95% while biopsy-derived positive predictive value (PPV3) for this category’s lesions remains low at 21.1% in the US. A major fallout of these facts is that we have way very high false positive rate leading to too many unnecessary biopsies and their associated costs and emotional burden. We improved our in-house intelligent-augmented Breast cancer RISK calculator (iBRISK), an integrated deep learning (DL) based decision support app and assessed its performance in a multicenter IRB-approved study. Methods: We improved iBRISK by retraining the DL model with an expanded dataset of 9,700 patient records of clinical risk-factors and mammographic descriptors from Houston Methodist Hospital (HMH) and validated using another 1,078 patient records. These patients were all seen between March 2006 and December 2016. We assessed the model using blinded, independent retrospective BI-RADS 4 patients who had biopsies subsequently after mammography and seen January 2015 - June 2019 at three major healthcare institutions in Texas, USA: MD Anderson Cancer Center, the University of Texas Health Sciences Center at San Antonio, and HMH. We dichotomized and trichotomized the data to evaluate precision of risk stratification and probability of malignancy (POM) estimation translated into biopsy decision augmentation. The iBRISK score as a continuous predictor of malignancy and possible cost savings was also analyzed. Results: The multicenter validation dataset had 4,209 women, median age (interquartile) was 56 (45, 65) years. The use of iBRISK score as a continuous predictor of malignancy yielded an AUC of 0.97. Among “low” and “moderate” POM patients, only two out of 1,228 patients (0.16%) and 118 out of 1788 (6.6%) were malignant respectively. This translates to an even better precision when compared to newly introduced BI-RADS 4 subcategories 4A and 4B, with associated PPV3s of 7.6% and 22%, respectively. The “high” POM group had a malignancy rate of 85.9% (1,025/1,193). Estimated potential cost savings in the US was over $260 million. Conclusion: The iBRISK app demonstrated high sensitivity in malignancy prediction and can potentially be used to safely obviate biopsies in up to 50% of patients in low/moderate POM-groups. This would result in significant healthcare quality improvement, cost savings, and help reduce patient anxiety. Citation Format: Chika F. Ezeana, Tiancheng He, Tejal A. Patel, Virginia Kaklamani, Maryam Elmi, Erica Ibarra, Pamela M. Otto, Kenneth A. Kist, Heather Speck, Lin Wang, Joe Ensor, Ya-Chen T. Shih, Bumyang Kim, I-Wen Pan, David Spak, Wei T. Yang, Jenny C. Chang, Stephen T. Wong. A multicenter study validated an integrated deep learning model for precision malignancy risk assessment and reducing unnecessary biopsies in BI-RADS 4 cases. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5698.

OBJECTIVE The objective of the study was to investigate the consequences of using C-arm cone-beam computed tomography (CBCT) on super-selective catheterization of hepatic artery (HA) branches during chemoembolization of hepatocellular carcinoma. METHODS Two groups of patients were created according to the dates of their treatment sessions. Group A and Group B included patients who had their treatment sessions in 2004 - 2005 and 2008 - 2010, respectively. The sessions performed in 2006 and 2007 were excluded to allow for the adoption and incorporation of CBCT imaging into clinical practice. All chemoembolized HA branches were categorized according to selection order (0-1, 2, or ≥3). Other procedure variables were documented. RESULTS A total of 58 and 183 sessions were included in Groups A and B, respectively, for 144 patients. C-arm CBCT was used in 2 (3%) sessions and 142 (78%) sessions in groups A and B, respectively. The average number of vessels treated was significantly higher in group B (1.8) compared to group A (1.3) (P < .0001). A shift to an increased selection order in group B (0-1, 44 [24%]; 2, 85 [46%]; ≥3, 54 [30%]) was more significant (P = .0004) than that in group A (0-1, 32 [55%]; 2, 18 [31%]; ≥3, 8 [14%]). The average duration of the procedure was significantly longer in group B (P = .0002). CONCLUSIONS Using C-arm CBCT during chemoembolization has a positive impact on increasing the number and order of HA selected and chemoembolized. This comes at the expense of an increase in the duration of the procedure.

Abstract Purpose: A Phase 2 trial of stereotactic radiotherapy and in situ cytotoxic virus therapy in patients with metastatic triple-negative breast cancer (mTNBC) followed by pembrolizumab (STOMP) was designed to evaluate dual approach of enhancing single-agent immune checkpoint blockade with adenovirus-mediated expression of herpes-simplex-virus thymidine-kinase (ADV/HSV-tk) plus valacyclovir gene therapy and stereotactic body radiotherapy (SBRT) in patients with mTNBC. Patients and Methods: In this single-arm, open-label Phase 2 trial, patients with mTNBC were treated with ADV/HSV-tk [5 × 1011 virus particles (vp)] intratumoral injection, followed by SBRT to the injected tumor site, then pembrolizumab (200 mg, every 3 weeks). The primary endpoint was clinical benefit rate [CBR; complete response (CR), partial response (PR), or stable disease (SD) ≥ 24 weeks per RECIST version1.1 at non-irradiated site]. Secondary endpoints included duration on treatment (DoT), overall survival (OS), and safety. Exploratory endpoints included immune response to treatment assessed by correlative tissue and blood-based biomarkers. Results: Twenty-eight patients were enrolled and treated. CBR was seen in 6 patients (21.4%), including 2 CR (7.1%), 1 PR (3.6%), and 3 SD (10.7%). Patients with clinical benefit had durable responses, with median DoT of 9.6 months and OS of 14.7 months. The median OS was 6.6 months in the total population. The combination was well tolerated. Correlative studies with Cytometry by Time of Flight (CyTOF) and imaging mass cytometry (IMC) revealed a significant increase of CD8 T cells in responders and of myeloid cells in non-responders. Conclusions: The median OS increased by more than 2-fold in patients with clinical benefit. The therapy is a well-tolerated treatment in heavily pretreated patients with mTNBC. Early detection of increased effector and effector memory CD8 T cells and myeloids correlate with response and non-response, respectively.

IntroductionImmunotherapy (IO) has transformed the treatment paradigm for a wide variety of solid tumours. However, assessment of response can be challenging with conventional radiological imaging (eg, iRECIST), which do not precisely capture the unique response patterns of tumours treated with IO. Emerging data suggest that circulating tumour DNA (ctDNA) can aid in response assessment in patients with solid tumours receiving IO. The short half-life of ctDNA puts it in a unique position for early treatment response monitoring. The BESPOKE IO study is designed to investigate the clinical utility of serial ctDNA testing to assess treatment response using a tumour-informed, bespoke ctDNA assay (Signatera) and to determine its impact on clinical decision-making with respect to continuation/discontinuation, or escalation/de-escalation of immunotherapy in patients with advanced solid tumours.Methods and analysisThe BESPOKE IO is a multicentre, prospective, observational study with a goal to enroll over 1500 patients with solid tumours receiving IO in up to 100 US sites. Patients will be followed for up to 2 years with serial ctDNA analysis, timed with every other treatment cycle. The primary endpoint is to determine the percentage of patients who will have their treatment regimen changed as guided by post-treatment bespoke ctDNA results along with standard response assessment tools. The major secondary endpoints include progression-free survival, overall survival and overall response rate based on the ctDNA dynamics.Ethics and disseminationThe BESPOKE IO study was approved by the WCG Institutional Review Board (Natera-20–043-NCP BESPOKE Study of ctDNA Guided Immunotherapy (BESPOKE IO)) on 22 February 2021. Data protection and privacy regulations will be strictly observed in the capturing, forwarding, processing and storing patients’ data. Natera will approve the publication of any study results in accordance with the site-specific contract.Trial registration numberNCT04761783.

1079 Background: Pembrolizumab, stereotactic body radiotherapy (SBRT), and viral vector-based gene therapy such as adenovirus-mediated expression of herpes simplex virus thymidine kinase (ADV/HSV-tk) plus ganciclovir have each shown antitumor immune activity. The combination of those modalities may represent a window of opportunity to enhance pembrolizumab efficacy in mTNBC patients. Methods: In this single-arm, open-label phase II trial, mTNBC patients were treated with in situ oncolytic ADV/HSV-tk (5 x 1011 vp) intratumoral injection, followed by SBRT to the injected tumor site, then pembrolizumab until progression or intolerable toxicity. Response was assessed in non-irradiated metastatic sites. The primary end point was clinical benefit rate (CR, PR and SD per RECIST version1.1). Secondary endpoints included duration on treatment (DoT) and safety. Immune correlative analysis with peripheral blood CYTOF (D1 baseline, D7 and D38), and tissue imaging mass cytometry (IMC) with 35 cell surface markers was performed on paired biopsies (D1 baseline and D17). Results: 28 mTNBC patients were enrolled; 18 (64.3%) had PD-L1 negative tumors. Median age was 54 years (range 34-78). Median prior lines of chemotherapy were 2 (range 0-6), with 8 (28.6%) having received >3. 3 (10.7%) had brain metastases. Clinical benefit was seen in 6 (21.43%) patients; 2 CR (7.1%),1 PR (3.57%) and 3 SD (10.7%). Patients who had clinical benefits had durable responses, with median DoT of 383 days (range 195-1195). One patient who had CR, but discontinued pembrolizumab due to Grade 3 pneumonitis, has remained disease free without any systemic therapy for 39 months. The combination was well tolerated; 9 (32.1%) patients had Grade 3- 4 AEs. Pre- and post-therapy CYTOF analysis showed significant association between immune biomarkers with clinical responses (AUC 0.75, Cohen’s Kappa 0.364). Tumor PDL1 was independently associated with response (AUC 0.70, Cohen’s Kappa 0.347); AUC for PDL1 together with immune biomarkers is 0.85. Conclusions: ADV/HSV-tk gene therapy followed by radiation therapy and then pembrolizumab is a well-tolerated promising treatment in heavily pretreated mTNBC patients. Early detection of increased effector and effector memory CD8 T cells and nonclassical monocytes correlates with response and non-response respectively. Clinical trial information: NCT03295916 .[Table: see text]