UH Seidman Cancer Center at ASTRO

Innovations in Cancer | Fall 2022

A large group of radiation oncology specialists and fellows from UH Seidman Cancer Center recently presented research findings and educational sessions at ASTRO. Here are the highlights:

Critical Appraisal of Data-Driven Studies: How Do We Evaluate the Cutting Edge in Artificial Intelligence, Real-World Data, Biomarkers and Comparative Effectiveness Studies?

Daniel Spratt, MD, Vincent K. Smith Chair in Radiation Oncology, provided education for ASTRO attendees on the process to develop, train, validate and commercialize biomarkers in oncology. He notes that prognostic biomarkers ideally have multiple prospective or post-hoc analyses of well-done clinical trials of large diverse cohorts with long-term follow-up, while predictive biomarkers ideally have at least one RCT validation of treatment of interest with significant biomarker-treatment interaction in relevant population. He concludes that AI biomarkers, especially if non-annotated/unbiased, require more rigorous data quality, quantity, diversity and independent validation.

Collaborative Approaches for Challenging CNS Cases: Neurosurgeons and Radiation Oncologists in Practice

Dr. Spratt also offered guidance to providers on how to setup a multidisciplinary spine oncology program and the optimal management of patients with spine metastasis. The best care of spine oncology patients, he says, is multidisciplinary care, involving radiation oncology, spine surgery, physical medicine and rehabilitation and interventional and diagnostic radiology. Standardization and rigorous QA/QC are also critical. Dr. Spratt concludes that large surgeries should be uncommonly done in metastatic patients and low-dose palliative radiation therapy should be selectively used as recurrence rates are high.

A Comprehensive Analysis of Institutional and Nationwide Datasets Identifies Key Parameters that Influence Racial Minority Enrollment onto Clinical Trials

Saad Sheikh, MD, and colleagues analyzed 352 cancer treatment trials from a multi-institutional cohort. They found that clinical trials treating breast and genitourinary malignancies had a favorable association with accruing a minority patient when compared to other solid tumors, while Phase II clinical trials had a less favorable association with accruing a minority patient when compared to early-phase trials. This analysis demonstrates that minority patients favorably associate with specific aspects of cancer treatment trials that should be addressed in future clinical trial design.

Biological Mechanisms of Radioresistance and Radiosensitivity with Traditional and Novel Systemic and Targeted Therapies in Breast Cancer

Corey Speers, MD, PhD, Hennessy Hyland Master Clinician in Immunotherapy and New Drug Development at UH Seidman Cancer Center, offered data and insights on the highly relevant clinical question of how novel systemic and targeted therapies for metastatic and non-metastatic breast cancer – including CDK4/6 inhibitors, capecitabine, poly ADP ribose polymerase (PARP) inhibitors, pertuzumab, ado-trastuzumab emtansine and pembrolizumab – impact the safety and efficacy of radiotherapy, discussing which systemic and targeted therapies sensitize or inhibit radiotherapy. He presented clinical data evaluating the safety and efficacy of radiotherapy administered concurrently with approved systemic therapies and detailed ongoing clinical trials and potential novel combination therapies.

Optimal Duration of Androgen Deprivation Therapy (ADT) with Definitive Radiotherapy for Prostate Cancer: An Individual Patient Data (IPD) Meta-Analysis from the International MARCAP Consortium

Nicholas Zaorsky, MD, and colleagues from the MARCAP consortium conducted an IPD meta-analysis performed using 13 randomized trials (EORTC 22863 and 22961, GICOR/DART, ICORG 9701, PCS III and IV, TROG 96.01 and RADAR, and NRG/RTOG 8610, 9202, 9408 and 9910). They found that for men enrolled in the context of randomized trials, after accounting for much of the non-compliance with durations of ADT beyond 18 months, 21 months is optimal for overall survival in men with intermediate- and high-risk prostate cancer. While longer durations of ADT provide incremental benefit on prostate cancer-specific survival, the researchers say this must be balanced against potential worsening of other-cause mortality, cost and quality of life.

A Survival Prediction Tool for Cancer Patients Treated with Stereotactic Radiation Therapy

Atallah Baydoun, MD, used data from the National Cancer Database to create a predictive model that identifies patients at greatest risk of death after receiving stereotactic radiotherapy as part of their cancer treatment. Data from 56,292 patients generated a predictive survival nomogram for cancer patients receiving stereotactic radiotherapy based on age, sex, morbidity, and site of metastases, radiation and primary tumor. Higher survival was associated with primary prostate or breast cancer (compared to lung or other primaries) and pelvis and breast radiation sites. Patients receiving stereotactic radiation to the brain and eye had the lowest median survival. The model was implemented as an online tool (https://tinyurl.com/rad-surv) that can provide improved decision-making process and treatment stratification.

Efficacy of Bicalutamide Monotherapy in Prostate Cancer: A Network Meta-Analysis of 10 Randomized Trials

Dr. Baydoun also reported results of the first-ever meta-analysis of bicalutamide monotherapy (BICmono) in prostate cancer. His results are strong evidence for the superiority of ADT over BICmono for men with prostate cancer. Importantly, Dr. Baydoun’s analysis showed that there was no improvement in outcomes from the use of BICmono compared to standard of care alone without the use ADT in localized prostate cancer. Thus, based on current randomized evidence, BICmono does not appear to have clinically meaningful activity in localized disease and standard of care alone should be recommended if a patient declines ADT.

Organized Social Medical Communication in Radiation Oncology: Two-Year Trends of Structured Hashtags

Last, Dr. Baydoun presented data on the use of radiation oncology hashtags on Twitter, such as: #Radonc, #PallOnc, #MedPhys, #SurvOnc, and #SuppOnc. This study is the first to report the influence of structured hashtags on radiation oncology Twitter communication. The two- year tracking data showed an increased popularity among healthcare providers, patients and organizations.

Patterns of Regional Radiotherapy Following Breast Conserving Surgery in Hormone Receptor Positive Breast Cancer with Limited Nodal Disease: a Single Institutional Experience

Eric Chen, MD, analyzed retrospective data from 2001-2021 from a single provider to eliminate inter-provider variation. Included in the analysis were patients who had estrogen receptor positive, HER2-negative breast cancer, and were found to have one to three positive axillary lymph nodes at time of upfront breast conserving surgery. Results over time show a significant shift in radiation target, with a decrease in treating breast alone in favor of treating the breast + low axilla. For patients in whom nodal irradiation was used, there was a significant increase in the inclusion of the internal mammary nodal chain. There was an increase in the utilization of sentinel lymph node biopsy resulting in a decrease in the median number of lymph nodes removed.

Impact of Hormone Receptor Status on Development of Brain Metastases from HER2+ Breast Cancer

Serah Choi, MD,PhD, and colleagues reviewed data on patients with brain metastases from breast cancer from a single academic institution from 2005–2020, during which time trastuzumab was included in the standard of care. Results show that of the 67 patients, 20 (29.9%) patients presented with one brain metastasis, 23 (34.3%) with two to nine brain metastases, 17 (25.4%) patients with ≥10 brain metastases, and seven (10.4%) with unknown number of brain metastases. The most salient factors associated with a decreased time interval to development of brain metastases included HR- status and presence of extracranial metastases at the time of initial HER2+ breast cancer diagnosis.

Incidence of Pneumonitis in Consolidative Durvalumab Era and its Correlation with Lung Dose-volumes Histograms in Stage III Nonresectable Non-small Cell Lung Cancer (NCSLC)

Tithi Biswas, MD, and colleagues conducted a retrospective study to compare the incidence of pneumonitis pre and post durvalumab era and its relationship with radiation dose parameters. Results show that there was a significant difference in the rate of symptomatic pneumonitis between chemoradiation alone (21%) versus chemoradiation with consolidative durvalumab (56%), suggesting that in comparison to chemoradiation alone, consolidative durvalumab was associated with significant increase in the rate of symptomatic pneumonitis. The research team says more studies are needed to understand the interaction between radiation parameters of lung and durvalumab.

Characteristics Impacting Overall Survival in Stage IV Non-small cell Lung Cancer Patients

Dr. Biswas and colleagues also undertook a retrospective analysis to assess the impact of modern day therapy on overall survival of stage IV NSCLC treated in a single institution over more than a decade. A total of 973 adults diagnosed with stage IV NSCLC were included in this analysis. Results show that the overall survival of stage IV NSCLC was significantly improved based on treatment between 2010 and 2015 and after 2015. As suggested from clinical trials, real world evidence suggests continued improvement in overall with improvement in systemic therapy, and other palliative measures. Additionally, having a histology of NSCLC NOS was significant for worse overall survival in this study.

Development of a Support Vector Machine-Based Tool for Survival Prediction after Stereotactic Body Radiotherapy for Liver Metastases Using Multi-Institutional Data

In addition, Dr. Biswas, medical student Kireem Nam and colleagues have used the Comprehensive Liver Pooled SBRT Outcomes (CaLiPSO) database, an international consortium of four high-volume SBRT academic centers, to identify patients who received radiation for liver metastases. Detailed demographic, clinical and dosimetric data were collected via chart review, and outcomes of interest were overall survival at 12, 18, and 24 months. Using clinical and dosimetric variables, the research team obtained promising results for accurate prediction of liver SBRT outcomes, making this the first SVM-based tool for SBRT outcome prediction in liver metastases validated using international, multi-institutional data for over 100 patients.

Efficacy and Toxicity of Stereotactic Body Radiation Therapy Treatment of Multiple Lung Lesions

Last, Dr. Biswas and colleagues conducted a retrospective study of how many courses of lung SBRT are safe and what factors predict toxicity. They compared dosimetry and clinical outcome parameters of these patients to a paired set of 50 randomly selected patients who received SBRT for a solitary lung lesion. Results suggest that SBRT treatment of multiple pulmonary tumors is safe and feasible without increasing in acute or late pulmonary toxicity. While there is significant difference in all lung parameters between these two groups, it did not translate into increasing toxicity.

Changes in Fractionation Schedules for Breast Cancer Treatment During the COVID-19 Pandemic

Courtney Pisano, MD, Janice Lyons, MD, and colleagues compared the use of ultrahypofractionated radiation (UHFR) schedules for breast cancer patients before COVID-19, at the height of the pandemic and after widespread vaccination and analyze factors predictive of use. They found that the use of UHFR increased during the post-vaccine period as compared to pre-COVID-19 and the pre-vaccine time periods (p<0.001), with 4.6%, 9.2% and 23%, receiving UHFR respectively. These data show that during COVID, there was a significant increase in the use of UHFR regimens for treatment of early-stage breast cancer patients that persisted and increased after the introduction of the vaccine despite return to pre-COVID patient volumes.

Synthetic FDG-Positron Emission Tomography Images for Patients with Non-Small Cell Lung Cancer: A Deep Learning-Based Approach Using Computed Tomography Images

Medical student Samar Bhat and colleagues conducted a retrospective study of patients undergoing stereotactic body radiotherapy for early-stage non-small cell lung cancer. For each patient, CT, measured PET images, and planning target (PTV) volumes were extracted using COMKAT Image Tool. Using an intelligent combination of windowed CT radiological features, the team obtained promising results for metabolic characteristic prediction. The achieved mean absolute error presents a considerable improvement in accuracy compared to the previously reported algorithm.