Neoadjuvant clinical trial designs: Challenges of the genomic era

Neoadjuvant Exercise Therapy in Prostate Cancer: A Phase 1, Decentralized Nonrandomized ControlledTrial

Importance

Observational data have shown that postdiagnosis exercise is associated with reduced risk of prostate cancer death. The feasibility and tumor biological activity of exercise therapy is not known.

Objective

To identify recommended phase 2 dose of exercise therapy for patients with prostate cancer.

Design, Setting, and Participants

This single-center, phase 1a dose-finding trial was conducted at a tertiary cancer center using a patientcentric, decentralized platform and included 53 inactive men with treatment-naive localized prostate cancer scheduled to undergo surgical resection between June 2019 and January 2023. Data were analyzed in June 2024.

Intervention

Six escalated exercise therapy dose levels ranging from 90 to 450 minutes per week of individualized, moderate-intensity treadmill walking, allocated using adaptive continual reassessment. All exercise therapy sessions were conducted remotely with real-time monitoring.

Main Outcomes and Measures

Feasibility was evaluated by relative exercise dose intensity (REDI). A dose level was considered feasible if 70% or more of patients achieved an REDI of 75% or greater. Activity end points were changes in tumor cell proliferation (Ki67) and plasma prostate-specific antigen levels between pretreatment and postintervention. Safety and changes in patient physiology were also assessed.

Results

A total of 53 men were enrolled (median [IQR] age, 61 [56-66] years). All dose levels were feasible (≥75% REDI). The mean (95% CI) changes in Ki67 were 5.0% (-4.3% to 14.0%) for 90 minutes per week, 2.4% (-1.3% to 6.2%) for 150 minutes per week, -1.3% (-5.8% to 3.3%) for 225 minutes per week, -0.2% (-4.0% to 3.7%) for 300 minutes per week, -2.6% (-9.2% to 4.1%) for 375 minutes per week, and 2.2% (-0.8% to 5.1%) for 450 minutes per week. Changes in prostate-specific antigen levels were 1.0 ng/mL (-1.8 to 3.8) for 90 minutes per week, 0.2 ng/mL (-1.1 to 1.5) for 150 minutes per week, -0.5 ng/mL (-1.2 to 0.3) for 225 minutes per week, -0.2 (-1.7 to 1.3) for 300 minutes per week, -0.7 ng/mL (-1.7 to 0.4) for 375 minutes per week, and -0.9 ng/mL (-2.4 to 0.7) for 450 minutes per week. No serious adverse events were observed. Overall, 225 minutes per week (approximately 45 minutes per treatment at 5 times weekly) was selected as the recommended phase 2 dose.

Conclusions and Relevance

The results of this nonrandomized clinical trial suggest that neoadjuvant exercise therapy is feasible and safe with promising activity in localized prostate cancer.

Trial Registration

ClinicalTrials.gov Identifier: NCT03813615.

Prior to Initiation of Chemotherapy, Can We Predict Breast Tumor Response? Deep Learning Convolutional Neural Networks Approach Using a Breast MRI Tumor Dataset

We hypothesize that convolutional neural networks (CNN) can be used to predict neoadjuvant chemotherapy (NAC) response using a breast MRI tumor dataset prior to initiation of chemotherapy. An institutional review board-approved retrospective review of our database from January 2009 to June 2016 identified 141 locally advanced breast cancer patients who (1) underwent breast MRI prior to the initiation of NAC, (2) successfully completed adriamycin/taxane-based NAC, and (3) underwent surgical resection with available final surgical pathology data. Patients were classified into three groups based on their NAC response confirmed on final surgical pathology: complete (group 1), partial (group 2), and no response/progression (group 3). A total of 3107 volumetric slices of 141 tumors were evaluated. Breast tumor was identified on first T1 postcontrast dynamic images and underwent 3D segmentation. CNN consisted of ten convolutional layers, four max-pooling layers, and dropout of 50% after a fully connected layer. Dropout, augmentation, and L2 regularization were implemented to prevent overfitting of data. Non-linear functions were modeled by a rectified linear unit (ReLU). Batch normalization was used between the convolutional and ReLU layers to limit drift of layer activations during training. A three-class neoadjuvant prediction model was evaluated (group 1, group 2, or group 3). The CNN achieved an overall accuracy of 88% in three-class prediction of neoadjuvant treatment response. Three-class prediction discriminating one group from the other two was analyzed. Group 1 had a specificity of 95.1% ± 3.1%, sensitivity of 73.9% ± 4.5%, and accuracy of 87.7% ± 0.6%. Group 2 (partial response) had a specificity of 91.6% ± 1.3%, sensitivity of 82.4% ± 2.7%, and accuracy of 87.7% ± 0.6%. Group 3 (no response/progression) had a specificity of 93.4% ± 2.9%, sensitivity of 76.8% ± 5.7%, and accuracy of 87.8% ± 0.6%. It is feasible for current deep CNN architectures to be trained to predict NAC treatment response using a breast MRI dataset obtained prior to initiation of chemotherapy. Larger dataset will likely improve our prediction model.