Clinical outcomes of melanoma brain metastases treated with nivolumab and ipilimumab alone versus nivolumab and ipilimumab with stereotactic radiosurgery.

PURPOSE: Upfront dual checkpoint blockade with immune checkpoint inhibitors (ICI) has demonstrated efficacy for treating melanoma brain metastases (MBM) in asymptomatic patients. Whether the combination of stereotactic radiosurgery (SRS) with dual checkpoint blockade improves outcomes over dual-checkpoint blockade alone is unknown. We evaluated clinical outcomes of patients with MBM receiving ICI with nivolumab and ipilimumab, with and without SRS. METHODS: 49 patients with 158 MBM receiving nivolumab and ipilimumab for untreated MBM between 2015 and 2022 were identified at our institution. Patient and tumor characteristics including age, Karnofsky Performance Status (KPS), presence of symptoms, cancer history, MBM burden, and therapy course were recorded. Outcomes measured from initiation of MBM-directed therapy included overall survival (OS), local control (LC), and distant intracranial control (DIC). Time-to-event analysis was conducted with the Kaplan-Meier method. RESULTS: 25 patients with 74 MBM received ICI alone, and 24 patients with 84 MBM received concurrent SRS. Median follow-up was 24 months. No differences in age (p = 0.96), KPS (p = 0.85), presence of symptoms (p = 0.79), prior MBM (p = 0.68), prior MBM-directed surgery (p = 0.96) or SRS (p = 0.68), MBM size (p = 0.67), or MBM number (p = 0.94) were seen. There was a higher rate of nivolumab and ipilimumab course completion in the SRS group (54% vs. 24%; p = 0.029). The SRS group received prior immunotherapy more often than the ICI alone group (54% vs. 8.0%; p < 0.001). There was no significant difference in 1-year OS (72% vs. 71%, p = 0.20) and DIC (63% v 51%, p = 0.26) between groups. The SRS group had higher 1-year LC (92% vs. 64%; p = 0.002). On multivariate analysis, LC was improved with combination therapy (AHR 0.38, p = 0.01). CONCLUSION: In our analysis, patients who received SRS with nivolumab and ipilimumab had superior LC without increased risk of toxicity or compromised immunotherapy treatment completion despite the SRS cohort having higher rates of prior immunotherapy. Further prospective study of combination nivolumab and ipilimumab with SRS is warranted.

Novel Definitive Hypofractionated Accelerated Radiation Dose-painting (HARD) for Unresected Soft Tissue Sarcomas.

Purpose: Soft tissue sarcomas (STS) are historically radioresistant, with surgery being an integral component of their treatment. With their low α/ß, STS may be more responsive to hypofractionated radiation therapy (RT), which is often limited by long-term toxicity risk to surrounding normal tissue. An isotoxic approach using a hypofractionated accelerated radiation dose-painting (HARD) regimen allows for dosing based on clinical risk while sparing adjacent organs at risk. Methods and Materials: We retrospectively identified patients from 2019 to 2022 with unresected STS who received HARD with dose-painting to high, intermediate, and low-risk regions of 3.0 Gy, 2.5 Gy, and 2.0 to 2.3 Gy, respectively, in 20 to 22 fractions. Clinical endpoints included local control, locoregional control, progression free survival, overall survival, and toxicity outcomes. Results: Twenty-seven consecutive patients were identified and had a median age of 68 years and tumor size of 7.0 cm (range, 1.2-21.0 cm). Tumors were most often high-grade (70%), stage IV (70%), located in the extremities (59%), and locally recurrent (52%). With a median follow-up of 33.4 months, there was a 3-year locoregional control rate of 100%. The 3-year overall and progression-free survival were 44.9% and 23.3%, respectively. There were 5 (19%) acute and 2 (7%) late grade 3 toxicities, and there were no grade 4 or 5 toxicities at any point. Conclusions: The HARD regimen is a safe method of dose-escalating STS, with durable 3-year locoregional control. This approach is a promising alternative for unresected STS, though further follow-up is required to determine long-term control and toxicity.

Clinical Outcomes of Patients with Non-Small Cell Lung Cancer Leptomeningeal Disease Following Receipt of EGFR-Targeted Therapy, Immune-Checkpoint Blockade, Intrathecal Chemotherapy, or Radiation Therapy Alone.

BACKGROUND: EGFR-targeted therapy (ETT) and immune-checkpoint blockade (ICB) have shown promising results in treating NSCLC brain metastases (BM). However, little is known of their effect in treating leptomeningeal disease (LMD). PATIENTS AND METHODS: This is a retrospective review of 80 patients diagnosed with NSCLC LMD from January 2014 to March 2021. Patients were grouped based on initial LMD treatment: radiotherapy (RT) alone, ETT, ICB, and intrathecal chemotherapy (ITC). RESULTS: EGFR mutation was present in 22 patients (28%). Twenty patients had positive cytology in cerebrospinal fluid, while 60 patients were diagnosed based on MRI with clinical correlation. The RT alone group consisted primarily of whole brain radiation (n = 20; 77%), stereotactic radiation (n = 3; 12%), and palliative spine radiation (n = 2; 7%). There were no significant differences amongst the treatment groups in age, performance status, or neurologic symptoms. Overall, the 6-month overall survival (OS) and craniospinal progression free survival (CS-PFS) were 35% and 24%, respectively. The 6-month OS for the ETT, ICB, ITC, and RT alone groups was 64%, 33%, 57%, and 29% respectively (log-rank P = .026). The 6-month CS-PFS for the ETT, ICB, ITC, and RT alone groups was 43%, 33%, 29%, and 19% respectively (log-rank P = .049). Upon univariate analysis, receipt of ETT compared to RT alone reached significance for OS (HR 0.35, P = .006) and CS-PFS (HR 0.39, P = .013). CONCLUSIONS: The prognosis for patients with NSCLC LMD remains poor overall. However, the receipt of ETT for patients with EGFR-positive disease was associated with improved outcomes.

Prediction of radiologic outcome-optimized dose plans and post-treatment magnetic resonance images: A proof-of-concept study in breast cancer brain metastases treated with stereotactic radiosurgery.

Background and purpose: Information in multiparametric Magnetic Resonance (mpMR) images is relatable to voxel-level tumor response to Radiation Treatment (RT). We have investigated a deep learning framework to predict (i) post-treatment mpMR images from pre-treatment mpMR images and the dose map ("forward models"), and, (ii) the RT dose map that will produce prescribed changes within the Gross Tumor Volume (GTV) on post-treatment mpMR images ("inverse model"), in Breast Cancer Metastases to the Brain (BCMB) treated with Stereotactic Radiosurgery (SRS). Materials and methods: Local outcomes, planning computed tomography (CT) images, dose maps, and pre-treatment and post-treatment Apparent Diffusion Coefficient of water (ADC) maps, T1-weighted unenhanced (T1w) and contrast-enhanced (T1wCE), T2-weighted (T2w) and Fluid-Attenuated Inversion Recovery (FLAIR) mpMR images were curated from 39 BCMB patients. mpMR images were co-registered to the planning CT and intensity-calibrated. A 2D pix2pix architecture was used to train 5 forward models (ADC, T2w, FLAIR, T1w, T1wCE) and 1 inverse model on 1940 slices from 18 BCMB patients, and tested on 437 slices from another 9 BCMB patients. Results: Root Mean Square Percent Error (RMSPE) within the GTV between predicted and ground-truth post-RT images for the 5 forward models, in 136 test slices containing GTV, were (mean ± SD) 0.12 ± 0.044 (ADC), 0.14 ± 0.066 (T2w), 0.08 ± 0.038 (T1w), 0.13 ± 0.058 (T1wCE), and 0.09 ± 0.056 (FLAIR). RMSPE within the GTV on the same 136 test slices, between the predicted and ground-truth dose maps, was 0.37 ± 0.20 for the inverse model. Conclusions: A deep learning-based approach for radiologic outcome-optimized dose planning in SRS of BCMB has been demonstrated.

Intricacies of Human-AI Interaction in Dynamic Decision-Making for Precision Oncology: A Case Study in Response-Adaptive Radiotherapy.

Background: Adaptive treatment strategies that can dynamically react to individual cancer progression can provide effective personalized care. Longitudinal multi-omics information, paired with an artificially intelligent clinical decision support system (AI-CDSS) can assist clinicians in determining optimal therapeutic options and treatment adaptations. However, AI-CDSS is not perfectly accurate, as such, clinicians' over/under reliance on AI may lead to unintended consequences, ultimately failing to develop optimal strategies. To investigate such collaborative decision-making process, we conducted a Human-AI interaction case study on response-adaptive radiotherapy (RT). Methods: We designed and conducted a two-phase study for two disease sites and two treatment modalities-adaptive RT for non-small cell lung cancer (NSCLC) and adaptive stereotactic body RT for hepatocellular carcinoma (HCC)-in which clinicians were asked to consider mid-treatment modification of the dose per fraction for a number of retrospective cancer patients without AI-support (Unassisted Phase) and with AI-assistance (AI-assisted Phase). The AI-CDSS graphically presented trade-offs in tumor control and the likelihood of toxicity to organs at risk, provided an optimal recommendation, and associated model uncertainties. In addition, we asked for clinicians' decision confidence level and trust level in individual AI recommendations and encouraged them to provide written remarks. We enrolled 13 evaluators (radiation oncology physicians and residents) from two medical institutions located in two different states, out of which, 4 evaluators volunteered in both NSCLC and HCC studies, resulting in a total of 17 completed evaluations (9 NSCLC, and 8 HCC). To limit the evaluation time to under an hour, we selected 8 treated patients for NSCLC and 9 for HCC, resulting in a total of 144 sets of evaluations (72 from NSCLC and 72 from HCC). Evaluation for each patient consisted of 8 required inputs and 2 optional remarks, resulting in up to a total of 1440 data points. Results: AI-assistance did not homogeneously influence all experts and clinical decisions. From NSCLC cohort, 41 (57%) decisions and from HCC cohort, 34 (47%) decisions were adjusted after AI assistance. Two evaluations (12%) from the NSCLC cohort had zero decision adjustments, while the remaining 15 (88%) evaluations resulted in at least two decision adjustments. Decision adjustment level positively correlated with dissimilarity in decision-making with AI [NSCLC: ρ = 0.53 ( p < 0.001); HCC: ρ = 0.60 ( p < 0.001)] indicating that evaluators adjusted their decision closer towards AI recommendation. Agreement with AI-recommendation positively correlated with AI Trust Level [NSCLC: ρ = 0.59 ( p < 0.001); HCC: ρ = 0.7 ( p < 0.001)] indicating that evaluators followed AI's recommendation if they agreed with that recommendation. The correlation between decision confidence changes and decision adjustment level showed an opposite trend [NSCLC: ρ = -0.24 ( p = 0.045), HCC: ρ = 0.28 ( p = 0.017)] reflecting the difference in behavior due to underlying differences in disease type and treatment modality. Decision confidence positively correlated with the closeness of decisions to the standard of care (NSCLC: 2 Gy/fx; HCC: 10 Gy/fx) indicating that evaluators were generally more confident in prescribing dose fractionations more similar to those used in standard clinical practice. Inter-evaluator agreement increased with AI-assistance indicating that AI-assistance can decrease inter-physician variability. The majority of decisions were adjusted to achieve higher tumor control in NSCLC and lower normal tissue complications in HCC. Analysis of evaluators' remarks indicated concerns for organs at risk and RT outcome estimates as important decision-making factors. Conclusions: Human-AI interaction depends on the complex interrelationship between expert's prior knowledge and preferences, patient's state, disease site, treatment modality, model transparency, and AI's learned behavior and biases. The collaborative decision-making process can be summarized as follows: (i) some clinicians may not believe in an AI system, completely disregarding its recommendation, (ii) some clinicians may believe in the AI system but will critically analyze its recommendations on a case-by-case basis; (iii) when a clinician finds that the AI recommendation indicates the possibility for better outcomes they will adjust their decisions accordingly; and (iv) When a clinician finds that the AI recommendation indicate a worse possible outcome they will disregard it and seek their own alternative approach.