BIOGUIDE-X: A First-in-Human Study of the Performance of Positron Emission Tomography-Guided Radiation Therapy

PURPOSE

Positron emission tomography (PET)-guided radiation therapy is a novel tracked dose delivery modality that uses real-time PET to guide radiation therapy beamlets. The BIOGUIDE-X study was performed with sequential cohorts of participants to (1) identify the fluorodeoxyglucose (FDG) dose for PET-guided therapy and (2) confirm that the emulated dose distribution was consistent with a physician-approved radiation therapy plan.

METHODS AND MATERIALS

This prospective study included participants with at least 1 FDG-avid targetable primary or metastatic tumor (2-5 cm) in the lung or bone. For cohort I, a modified 3 + 3 design was used to determine the FDG dose that would result in adequate signal for PET-guided therapy. For cohort II, PET imaging data were collected on the X1 system before the first and last fractions among patients undergoing conventional stereotactic body radiation therapy. PET-guided therapy dose distributions were modeled on the patient's computed tomography anatomy using the collected PET data at each fraction as input to an "emulated delivery" and compared with the physician-approved plan.

RESULTS

Cohort I demonstrated adequate FDG activity in 6 of 6 evaluable participants (100.0%) with the first injected dose level of 15 mCi FDG. In cohort II, 4 patients with lung tumors and 5 with bone tumors were enrolled, and evaluable emulated delivery data points were collected for 17 treatment fractions. Sixteen of the 17 emulated deliveries resulted in dose distributions that were accurate with respect to the approved PET-guided therapy plan. The 17th data point was just below the 95% threshold for accuracy (dose-volume histogram score = 94.6%). All emulated fluences were physically deliverable. No toxicities were attributed to multiple FDG administrations.

CONCLUSIONS

PET-guided therapy is a novel radiation therapy modality in which a radiolabeled tumor can act as its own fiducial for radiation therapy targeting. Emulated therapy dose distributions calculated from continuously acquired real-time PET data were accurate and machine-deliverable in tumors that were 2 to 5 cm in size with adequate FDG signal characteristics.

Surrogate endpoints in clinical trials of p16-positive squamous cell carcinoma of the oropharynx: an individual patient data meta-analysis

BACKGROUND

The increased incidence of human papillomavirus (HPV)-related cancers has motivated efforts to optimise treatment for these patients with excellent prognosis. Validation of surrogates for overall survival could expedite the investigation of new therapies. We sought to evaluate candidate intermediate clinical endpoints in trials assessing definitive treatment of p16-positive oropharyngeal cancer with chemotherapy or radiotherapy.

METHODS

We did a retrospective review of five multicentre, randomised trials (NRG/RTOG 9003, 0129, 0234, 0522, and 1016) that tested radiotherapy with or without chemotherapy in patients (aged ≥18 years) with p16-positive localised head or neck squamous-cell carcinomas. Eight intermediate clinical endpoints were considered as potential surrogates for overall survival: freedom from local progression, freedom from regional progression, freedom from distant metastasis, freedom from locoregional progression, freedom from any progression, locoregional progression-free survival, progression-free survival, and distant metastasis-free survival. We used a two-stage meta-analytical framework, which requires high correlation between the intermediate clinical endpoint and overall survival at the patient level (condition 1), and high correlation between the treatment effect on the intermediate clinical endpoint and the treatment effect on overall survival (condition 2). For both, an r2 greater than 0·7 was used as criteria for clinically relevant surrogacy.

FINDINGS

We analysed 1373 patients with oropharyngeal cancer from May 9, 2020, to Nov 22, 2023. 1231 (90%) of patients were men, 142 (10%) were women, and 1207 (88%) were White, with a median age of 57 years (IQR 51-62). Median follow-up was 4·2 years (3·1-5·1). For the first condition, correlating the intermediate clinical endpoints with overall survival at the individual and trial level, the three composite endpoints of locoregional progression-free survival (Kendall's τ 0·91 and r2 0·72), distant metastasis-free survival (Kendall's τ 0·93 and r2 0·83), and progression-free survival (Kendall's τ 0·88 and r2 0·70) were highly correlated with overall survival at the patient level and at the trial-group level. For the second condition, correlating treatment effects of the intermediate clinical endpoints and overall survival, the composite endpoints of locoregional progression-free survival (r2 0·88), distant metastasis-free survival (r2 0·96), and progression-free survival (r2 0·92) remained strong surrogates. Treatment effects on the remaining intermediate clinical endpoints were less strongly correlated with overall survival.

INTERPRETATION

We identified locoregional progression-free survival, distant metastasis-free survival, and progression-free survival as surrogates for overall survival in p16-positive oropharyngeal cancers treated with chemotherapy or radiotherapy, which could serve as clinical trial endpoints.

FUNDING

NRG Oncology Operations, NRG Oncology SDMC, the National Cancer Institute, Eli Lilly, Aventis, and the University of Michigan.

Lessons Learned from a Multi-Site, Team-Based Serious Illness Care Program Implementation at an Academic Medical Center

Background: Patients with serious illness benefit from conversations to share prognosis and explore goals and values. To address this, we implemented Ariadne Labs' Serious Illness Care Program (SICP) at Stanford Health Care. Objective: Improve quantity, timing, and quality of serious illness conversations. Methods: Initial implementation followed Ariadne Labs' SICP framework. We later incorporated a team-based approach that included nonphysician care team members. Outcomes included number of patients with documented conversations according to clinician role and practice location. Machine learning algorithms were used in some settings to identify eligible patients. Results: Ambulatory oncology and hospital medicine were our largest implementation sites, engaging 4707 and 642 unique patients in conversations, respectively. Clinicians across eight disciplines engaged in these conversations. Identified barriers that included leadership engagement, complex workflows, and patient identification. Conclusion: Several factors contributed to successful SICP implementation across clinical sites: innovative clinical workflows, machine learning based predictive algorithms, and nonphysician care team member engagement.

Individualized Stereotactic Ablative Radiotherapy for Lung Tumors: The iSABR Phase 2 Nonrandomized Controlled Trial

Importance

Stereotactic ablative radiotherapy (SABR) is used for treating lung tumors but can cause toxic effects, including life-threatening damage to central structures. Retrospective data suggested that small tumors up to 10 cm3 in volume can be well controlled with a biologically effective dose less than 100 Gy.

Objective

To assess whether individualizing lung SABR dose and fractionation by tumor size, location, and histological characteristics may be associated with local tumor control.

Design, Setting, and Participants

This nonrandomized controlled trial (the iSABR trial, so named for individualized SABR) was a phase 2 multicenter trial enrolling participants from November 15, 2011, to December 5, 2018, at academic medical centers in the US and Japan. Data were analyzed from December 9, 2020, to May 10, 2023. Patients were enrolled in 3 groups according to cancer type: initial diagnosis of non-small cell lung cancer (NSCLC) with an American Joint Committee on Cancer 7th edition T1-3N0M0 tumor (group 1), a T1-3N0M0 new primary NSCLC with a history of prior NSCLC or multiple NSCLCs (group 2), or lung metastases from NSCLC or another solid tumor (group 3).

Intervention

Up to 4 tumors were treated with once-daily SABR. The dose ranged from 25 Gy in 1 fraction for peripheral tumors with a volume of 0 to 10 cm3 to 60 Gy in 8 fractions for central tumors with a volume greater than 30 cm3.

Main outcome

Per-group freedom from local recurrence (same-lobe recurrence) at 1 year, with censoring at time of distant recurrence, death, or loss to follow-up.

Results

In total, 217 unique patients (median [IQR] age, 72 [64-80] years; 129 [59%] male; 150 [69%] current or former smokers) were enrolled (some multiple times). There were 240 treatment courses: 79 in group 1, 82 in group 2, and 79 in group 3. A total of 285 tumors (211 [74%] peripheral and 74 [26%] central) were treated. The most common dose was 25 Gy in 1 fraction (158 tumors). The median (range) follow-up period was 33 (2-109) months, and the median overall survival was 59 (95% CI, 49-82) months. Freedom from local recurrence at 1 year was 97% (90% CI, 91%-99%) for group 1, 94% (90% CI, 87%-97%) for group 2, and 96% (90% CI, 89%-98%) for group 3. Freedom from local recurrence at 5 years ranged from 83% to 93% in the 3 groups. The proportion of patients with grade 3 to 5 toxic effects was low, at 5% (including a single patient [1%] with grade 5 toxic effects).

Conclusions and Relevance

The results of this nonrandomized controlled trial suggest that individualized SABR (iSABR) used to treat lung tumors may allow minimization of treatment dose and is associated with excellent local control. Individualized dosing should be considered for use in future trials.

Trial Registration

ClinicalTrials.gov Identifier: NCT01463423.

First-Year Experience of IMRT/SBRT Treatments Using a Novel Biology-Guided Radiation Therapy System

PURPOSE/OBJECTIVE(S)

This study presents the first-year experience of treating patients using intensity-modulated radiation therapy (IMRT) and stereotactic body radiation therapy (SBRT) with the X1 system, the first biology-guided radiation therapy (BgRT) machine installed in a clinical setting.

MATERIALS/METHODS

A total of 78 patients underwent IMRT and SBRT treatments on the X1 system from May 2021 to May 2022. Clinical and technical data, such as treatment sites, number of pre-treatments kVCT scans, beam on time, patient setup time, imaging time per kVCT, and couch shifts after kVCT match, were collected and analyzed. Additionally, daily machine output stability, patient-specific quality assurance (QA) results, machine uptime, and user survey were also documented and reported.

RESULTS

The most commonly treated site was the head and neck (63%), followed by the pelvis (23%), thorax (6%), and abdomen (8%). All treatments, except for 5 pelvis patients (6%) who received SBRT treatments for bony metastases, were conventionally fractionated IMRT (CF IMRT). The average number of kVCT scans per fraction is 1.2 ± 0.5 for all treatments. The average beam on time in minutes was 9.2 ± 3.5 for all treatments, 8.4 ± 2.4 for head and neck, 6.7 ± 1.3 for thorax, 10.3 ± 1.6 for abdomen, 11.6 ± 5.1 for CF IMRT pelvis, and 10.8 ± 5.3 for SBRT pelvis. The average patient setup time and imaging time per kVCT was 4.8 ± 2.6 minutes and 4.6 ± 1.5 minutes, respectively. The average couch corrections based on kVCT images were 0.4 ± 4.4 mm, 1.0 ± 4.5 mm, and 1.3 ± 4.3 mm along the x, y, and z direction, respectively; the average couch rotation corrections were 0.1 ± 0.9° for pitch, 0.0 ± 0.9° for roll, and 0.2 ± 1.2° for yaw. The daily machine output was 0.4 ± 1.2% from the baseline. The patient QA had a gamma passing rate of 97.4 ± 2.8%. The machine uptime was 92% of the total treatment time. The kVCT image quality and daily QA process received the highest level of satisfaction, while the treatment workflow for therapists received the lowest level of satisfaction (table 1).

CONCLUSION

At one year after the installation of the X1 system, this study reports successful treatment of 78 patients using IMRT/ SBRT. With the recent FDA clearance of BgRT, our institution is preparing to treat patients using PET-guidance via a new product release, which should address deficiencies in the current IGRT workflow.

Reinforcement Learning Powered Station Parameter Optimized Radiation Therapy (SPORT): A Novel Treatment Planning and Beam Delivery Technique

PURPOSE/OBJECTIVE(S)

Conventional intensity modulated radiation therapy (IMRT) with a typical 5-20 fixed beams often does not provide sufficient angular sampling required for conformal dose shaping, whereas current volumetric modulated arc therapy (VMAT) discretizes the angular space into equally spaced control points without considering the differential need for intensity modulation of different angles, leading to undersampling at some angles while oversampling at some other angles. Our goal is to develop a node or station parameter optimized radiation therapy (SPORT) strategy with simultaneously optimized angular sampling and beam modulation by leveraging state-of-the-art reinforcement learning and the unique capability of modern digital LINACs in dose delivery through programmable nodal points.

MATERIALS/METHODS

We developed a SPORT optimization framework, in which, the process of programming control points (or station parameters) was formulated as a stochastic dynamic programming problem, which was solved by a reinforcement learning-based algorithm. On-policy reinforcement learning method, namely, state-action-reward-state-action (SARSA) was integrated with deep convolutional neural network to predict station parameters by utilizing the patient's anatomical structures meanwhile considering the delivery capability of a typical digital LINAC machine. Here, the deep convolutional neural network estimated the state-action value by using the quality of the plan with current station parameters when a next potential station parameter was selected. The state-action value was then updated by SARSA learning. The quality of the plan was quantified by dosimetry constraints. The model was assessed by a retrospective study on a cohort of patients underwent head-and-neck radiation therapy. Dosimetric analysis and delivery efficiency comparisons were used to evaluate the performance of the proposed framework.

RESULTS

Our model was used to generate 16 plans unseen in the original training set. All the plans predicted by our model achieved better dose distributions without violating clinical planning constraints. Moreover, instead of using 4 full standard arcs in the original clinically used plans obtained via manual optimization, the predicted plans only used one full standard arc (about 178 control points) plus boost from a few sub-arcs (less than 30 degrees of gantry angles), which significantly improved the efficiency of the beam delivery. We are in the process of integrating the sub-arcs into the full arc by considering the programmable capability of modern LINACs.

CONCLUSION

We demonstrated that a machine learning-based SPORT framework capable of optimizing the spatial sampling and beam modulation simultaneously for modern radiation therapy. The framework not only significantly improves the quality and efficiency of beam delivery, but also has the potential to be incorporated into current clinical workflow to improve the efficiency of dose planning and delivery.

Patient-Specific Auto-segmentation on Daily kVCT Images for Adaptive Radiation Therapy

PURPOSE

This study explored deep-learning-based patient-specific auto-segmentation using transfer learning on daily RefleXion kilovoltage computed tomography (kVCT) images to facilitate adaptive radiation therapy, based on data from the first group of patients treated with the innovative RefleXion system.

METHODS AND MATERIALS

For head and neck (HaN) and pelvic cancers, a deep convolutional segmentation network was initially trained on a population data set that contained 67 and 56 patient cases, respectively. Then the pretrained population network was adapted to the specific RefleXion patient by fine-tuning the network weights with a transfer learning method. For each of the 6 collected RefleXion HaN cases and 4 pelvic cases, initial planning computed tomography (CT) scans and 5 to 26 sets of daily kVCT images were used for the patient-specific learning and evaluation separately. The performance of the patient-specific network was compared with the population network and the clinical rigid registration method and evaluated by the Dice similarity coefficient (DSC) with manual contours being the reference. The corresponding dosimetric effects resulting from different auto-segmentation and registration methods were also investigated.

RESULTS

The proposed patient-specific network achieved mean DSC results of 0.88 for 3 HaN organs at risk (OARs) of interest and 0.90 for 8 pelvic target and OARs, outperforming the population network (0.70 and 0.63) and the registration method (0.72 and 0.72). The DSC of the patient-specific network gradually increased with the increment of longitudinal training cases and approached saturation with more than 6 training cases. Compared with using the registration contour, the target and OAR mean doses and dose-volume histograms obtained using the patient-specific auto-segmentation were closer to the results using the manual contour.

CONCLUSIONS

Auto-segmentation of RefleXion kVCT images based on the patient-specific transfer learning could achieve higher accuracy, outperforming a common population network and clinical registration-based method. This approach shows promise in improving dose evaluation accuracy in RefleXion adaptive radiation therapy.

Clinical Implementation of an Automated IMRT/VMAT Treatment Planning Tool

PURPOSE/OBJECTIVE(S)

To create an in-house automated treatment planning tool for IMRT/VMAT treatments and evaluate the dosimetric plan quality against manually generated plans.

MATERIALS/METHODS

A scripting application programming interface is employed to interact with a commercial treatment planning system (TPS) to implement automatic plan evaluation and update optimization parameters by mimicking the human planning process. The automated planning performs in an iterative fashion until reaching an acceptable tradeoff among target coverage/dose homogeneity and sparing of critical organs at risk. In each iteration, the dose constraints, priorities, and optimization structures for are automatically updated based on the results of the current iteration. Twenty previously treated plans (10 prostate and 10 head and neck), were preliminarily used to evaluate the performance of the automated planning tool. The differences in target and organ-at-risk metrics from the manually generated clinical plans were analyzed using paired t-test to evaluate clinical acceptability of tour automated planning tool. The current in-house-developed automated planning solution is able to create plans for different disease sites, including head & neck, prostate, pelvis, and lung. So far, the VMAT plans for more than 150 different cases have been generated with the tool. The results for these were also evaluated.

RESULTS

Compared to the manually generated clinical head and neck plans, all auto plans achieved PTV D95% coverage and critical organs at risk sparing without statistically significant change in average global Dmax (107.4% for manual vs 107.3% for automated plans). The auto-planning solution provided reduced maximum doses to brainstem and spinal cord (average reductions with standard deviations of 5.1 ± 2.6 Gy and 2.9 ± 1.4 Gy, respectively, all p <0.03), reduced average mean doses to contralateral parotid, ipsilateral parotid, contralateral submandibular gland, pharynx, esophagus, cochleae (reductions of 2.2 ± 2.9 Gy, 4.8 ± 4.7 Gy, 3.6 ± 5.2 Gy, 2.0 ± 7.1 Gy, 3.9 ± 2.6 Gy, 3.8 ± 5.0 Gy, respectively, all p < 0.045). Similar results were observed for the prostate plans. With the same PTV coverage and without statistically significant change in average global Dmax (106.5% for manual vs 106.8% for automated plans), the automated solution provided superior sparing for both bladder and rectum. Bladder V75, V70, V65 were reduced by 0.6% ± 2.1%, 0.8% ± 2.5%, and 0.9% ± 2.9% (all p <0.04), respectively. Rectum V75, V70, V65, V60 were reduced by 1.0% ± 2.3%, 1.2% ± 2.8%, 1.3% ± 3.2%, 1.6% ± 3.6% (all p < 0.01), respectively.

CONCLUSION

Our automated treatment planning solution is capable of efficiently generating VMAT plans for different disease sites with superior dosimetric indices compared to manually generated plans. Our tool is integrated within a commercial TPS platform, so it has the advantage of seamless adoption into the standard workflow to improve plan quality and treatment planning efficiency in our clinic.

Patient-Adaptive Automated Segmentation in Daily kVCT Images for Radiotherapy of Head and Neck and Prostate Cancer

PURPOSE/OBJECTIVE(S)

The purpose of this study was to examine the use of transfer learning in deep learning-based auto-segmentation of daily kilovoltage computed tomography (kVCT) images for patient-specific adaptive radiotherapy. Using data from the first cohort of patients treated with the innovative BgRT system, the objective of this study was to evaluate the potential benefits of this approach in facilitating efficient and effective adaptive radiotherapy.

MATERIALS/METHODS

For the head and neck (HaN) site and pelvic site, we first trained a deep convolutional segmentation network using a population dataset, consisting of 67 and 56 patient cases, respectively. This population network was then fine-tuned for a specific patient using a transfer learning approach, adapting the network weights. The auto-segmentation network utilized in this study was a 23-layer U-Net with batch normalization, a dropout rate of 0.5, and four skip connections between the encoder and decoder at different levels. We used initial planning CT and 5-26 sets of daily kVCT scans with a total of 8,039 images for patient-specific learning in the 6 HaN cases and 4 pelvic cases, particularly analyzing the relationship between the number of sequential patient-specific training data and the performance of the auto-segmentation. We compared the performance of the patient-specific network with the population network and the clinical rigid registration method, using the Dice similarity coefficient (DSC) as the evaluation metric. Additionally, we investigated the corresponding dosimetric impacts of the different auto-segmentation and registration methods.

RESULTS

The patient-specific network showed improved mean DSC scores of 0.88 and 0.90 for three HaN organs at risk (OARs) and eight pelvic targets and OARs, respectively, compared to the population network (0.70 and 0.63) and the registration method (0.72 and 0.72). The DSC of the patient-specific network steadily improved as the number of longitudinal training cases increased, reaching near saturation after 6 training cases. The use of the patient-specific auto-segmentation resulted in a reduction of the mean discrepancy in target and OAR doses between delivery and planning from 5.5% with the clinical rigid registration to 1.1%.

CONCLUSION

The use of patient-specific transfer learning in auto-segmenting kVCT images showed higher accuracy compared to a conventional population network and clinical registration-based method. This approach holds promise for enhancing dose evaluation accuracy in adaptive radiotherapy.

Analysis of the Measured FDG Uptake from the First-in-Human Clinical Trial of Biology-Guided Radiotherapy

PURPOSE/OBJECTIVE(S)

The RefleXion X1 system is a novel linear accelerator equipped with dual 90° PET arcs incorporated into its architecture to capture emissions from tumors and designed to respond by directing the radiation beam towards target. This study reports on the measured FDG uptake from the first in human multi-institutional clinical trial (BIOGUIDE-X) evaluating the performance and safety of the RefleXion X1 PET-LINAC.

MATERIALS/METHODS

A total of nine patients treated with stereotactic body radiotherapy (SBRT) for lung (5) and bone (4) tumors were enrolled in the Cohort II of this study after screening their pre-study diagnostic PET/CT, acquired up to 60 days prior to enrollment, to ensure their tumor size between 2 to 5 cm and SUVmax >6. After CT simulation, the tumor and OARs were delineated, and patients had a 4-pass Imaging-only (BgRT Modeling) PET/CT acquisition on the X1 system to generate biology-guided radiotherapy (BgRT) plans. Before the patients' first and last SBRT fractions, they were injected with FDG, and short PET pre-scan (1-pass) was performed on the X1 followed by a long-PET acquisition (4-pass) to emulate the expected BgRT dose distribution without firing beam. Patients were also imaged on a third-party diagnostic PET/CT scanner after the last-fraction X1 scan. This study compares the SUVmax from the screening PET/CT, X1 Imaging-only scan, X1 PET pre-scan and long scan before the first and last-fractions, and final diagnostic PET/CT.

RESULTS

The median time from injection to PET imaging was 84 ± 15.4 mins for X1 Imaging-only (used for generating BgRT plans), 77 ± 21.6 mins for X1 pre-scan (safety check before treatment start), 108+/- 22 mins for X1 long-PET (used to emulate treatment delivery), and 161 ± 23 mins for final diagnostic PET. For a nominal 10 mCi injection, the mean SUVmax for screening imaging performed on the diagnostic PET/CT was 10.8 ± 4.3. For a 15 mCi nominal injection, the mean SUVmax calculated on the X1 was 5.3 ± 2.6, 5.4 ± 2.0, 5.5 ± 2.6, 5.2 ± 1.8 and 5.4 ± 2.2 for the Imaging-only, first-fraction PET pre-scan, first-fraction long PET scan, last-fraction PET pre-scan, and last-fraction long PET scan, respectively. The overall median SUVmax for all patients across all timepoints and scans with X1 was calculated to be 4.8 with a range of 2.4 to 9.8. The median SUVmax for the diagnostic PET/CT scan after the last fraction X1 scan was 15.8 with a range of 8.5 to 27.7.

CONCLUSION

The dual PET arcs and limited axial extent of the X1 PET subsystem results in lower system sensitivity in comparison to diagnostic PET scanners equipped with full ring and larger axial extent, as expected. With the same FDG injection, the RefleXion X1 produced SUVmax values that were 30.4 % of the diagnostic PET/CT scanners' values. Nevertheless, the X1 collected sufficient emission data to enable successful completion of emulated BgRT deliveries that met dose accuracy criteria in a clinical setting.

Imaging Performance of the PET Scan on a Novel Ring Gantry-Based PET/CT Linear Accelerator System in the First-in-Human Study of Biology-Guided Radiotherapy

PURPOSE/OBJECTIVE(S)

Biology-guided radiotherapy (BgRT) is a novel tracked dose delivery modality using real-time positron emission tomography (PET) to guide radiotherapy beamlets. The present study was performed with sequential cohorts of participants to evaluate the performance and safety of BgRT. Primary endpoints were previously reported. We hereby report on one of the secondary endpoints assessing a novel treatment planning machine with integrated dual kVCT/PET imaging ("novel device") performance in comparison to a third-party diagnostic PET/CT scan.

MATERIALS/METHODS

This single-arm, open-label, prospective study included participants with at least 1 FDG-avid targetable primary or metastatic tumor (≥2cm and ≤5cm) in the lung or bone. PET imaging data were collected on the novel device and on a third-party diagnostic PET/CT performed in sequence once at the planning timepoint in Cohort I, and immediately before the last fraction among patients undergoing stereotactic radiotherapy in Cohort II. Three central read radiation oncologists (CRRO) provided an interpretation of the novel device PET scans which were compared to an agreement standard based on 3 central radiologists' review of the paired diagnostic PET/CT scan. Positive percent agreement for localization of the target tumor within the biology-tracking zone (BTZ) was the key metric because it reflects whether advancing patients to subsequent steps in the BgRT workflow based on the novel device's imaging was ultimately appropriate.

RESULTS

In Cohort 1, 6 image comparisons were performed. The positive (%) agreement for the aggregate radiation oncologist review was 100% (5/5), reflecting that in all 5 cases where the aggregate radiation oncologists deemed the tumor to fall within the BTZ based upon the novel device PET images, the central radiologists came to the same conclusion upon review of the paired diagnostic PET/CT images. The overall (%) agreement for the aggregate radiation oncologist review was 83.3% (5/6): localization was not established on the novel device in 1 case, even though it was established on the diagnostic PET/CT. This would not pose risk in real world practice as BgRT candidacy would be aborted for tumors not visible on the novel device. In Cohort II, among the 7 image comparisons, there was 100% positive percent agreement between the aggregate CRRO and the agreement standard as the localization criteria was met in both scans for all 7 patients. This was concordant with a 100% overall percent agreement.

CONCLUSION

This investigation demonstrated a 100% positive percent agreement between central review of this novel device images by radiation oncologists and central review of the accompanying third-party PET/CT images by radiologists. There were no cases where a positive localization by the aggregate CRRO was not confirmed by the third-party PET/CT standard, providing evidence against the likelihood of falsely positive localizations on the novel device that would inappropriately advance patients in the workflow.

Workflow Considerations for Biology-Guided Radiotherapy (BgRT) Implementation

PURPOSE/OBJECTIVE(S)

Biology-guided radiotherapy (BgRT) is a novel platform that combines real-time PET imaging with a 6MV Linac to target tumors. The performance and safety of BgRT was assessed in the BIOGUIDE-X clinical trial. This study aims to report on the BgRT workflow steps and assess the time required for each step of the BgRT process during this trial.

MATERIALS/METHODS

A total of nine patients were enrolled in the second Cohort of the BIOGUIDE-X study which included patients treated with stereotactic body radiotherapy (SBRT) for lung tumors (5) and bone tumors (4). The pre-treatment BgRT workflow includes CT simulation, contouring, imaging-only (BgRT Modeling) PET acquisition, BgRT planning, patient specific QA and plan approval. The imaging-only PET acquisition on the X1 collects a representative PET volumetric 3D image and is an input to develop the BgRT treatment plan. The steps during the BgRT delivery session are kVCT localization, PET pre-scan, PET evaluation and BgRT delivery. The PET PreScan is a 1-pass short-duration PET acquisition that is used to confirm that the PET biodistribution on the day of treatment is consistent with that of the imaging-only PET. During BIOGUIDE-X, the BgRT delivery step was replaced by a 4-pass long-PET acquisition that was used to emulate the expected BgRT dose distribution without turning the beam on. To assess BgRT workflow, times from 18F-FDG injection to image-only PET acquisition, 18F-FDG injection to PET pre-scan, Pre-scan to PET evaluation, and PET evaluation to BgRT delivery (long PET acquisition) were recorded.

RESULTS

Time between the 18F-FDG injection and the X1 imaging-only PET scan was 84 ± 19 minutes which includes time for 18F-FDG update. Average time to perform imaging-only PET scan was 26 ± 4 minutes. During the BgRT 'delivery' session, the mean time between the kVCT acquisition and PET pre-scan acquisition was 7 ± 3 minutes. The mean time to acquire a 1-pass PET pre-scan was 6 ± 1 then followed by 6 ± 1 minutes for the PET pre-scan dose calculation to estimate the BgRT doses that it would have delivered for this fraction. On average, the PET reconstruction, the PET signal localization verification and the evaluation of safety metrics took 11 ± 4 minutes. The mean time for BgRT 'delivery' was 27 ± 5 minutes based on the 4-pass long PET acquisition. Time from the start of the BgRT session to the end of the BgRT 'delivery' with this version of the investigative product release was 65 ± 9 minutes.

CONCLUSION

The new processes introduced by the BgRT technology were evaluated and found clinically feasible. Improvements are being undertaken to shorten the time required for each step and to increase patient comfort ahead of BgRT clinical implementation.

Predicting Local Control with Dosimetric Parameters in Patients Receiving Individualized Stereotactic Ablative Radiotherapy for Lung Tumors

PURPOSE/OBJECTIVE(S)

Stereotactic ablative radiotherapy (SABR) is an effective treatment option for lung tumors. The individualized lung tumor SABR (iSABR) trial was a phase II single-arm study that personalized lung tumor SABR dose and fractionation based on tumor size, location, and histology with very low rates of local recurrence (LR). A secondary analysis of this trial was conducted to assess for potential dosimetric predictors of LR, in order to help guide future clinical treatment planning.

MATERIALS/METHODS

From 2011 to 2018, local, regional and distant recurrence data were prospectively collected from 204 patients (261 lung SABR treatments) enrolled in a prospective trial. Baseline characteristics and treatment details were evaluated. Dosimetric and treatment plan parameters were evaluated for their potential to predict LR, using logistic regression and chi-squared analyses.

RESULTS

The majority of treated tumors were peripheral (71%, vs 29% central), primary lesions (76%, versus 24% metastatic), and of adenocarcinoma histology (67%, versus 13% squamous cell carcinoma and 19% other). The median follow-up was 24 months (range 2-95). Twenty-seven (10.3%) LRs occurred, with a median time to LR of 15 months (range 6-81 months). There were no significant associations between the overall cohort and the dosimetric parameters. However, for the multi-fraction cohort, an increased proportion of the PTV receiving 110% and 115% of the prescription dose were associated with lower LR (p = 0.01 and p = 0.01 respectively). Specifically for the 50 Gy in 4 fraction cohort, an increased D1cc, D0.03cc, as well as the proportion of the PTV receiving 110%, 115%, and 120% of the prescription dose were associated with lower LR (p < 0.001, p = 0.001, p = 0.003, p < 0.001, p = 0.004, respectively). There was no association of LR with prescription dose expressed as biologically effective dose using an alpha/beta of 10 Gy (BED10), D99%, or single- versus multi-fraction regimens.

CONCLUSION

SABR for lung tumors using the individualized protocol on this trial showed excellent LR rates. We identified dosimetric parameters that were associated with LR, including V110% and V115% within the multi-fraction cohort, as well as the 50 Gy in 4 fraction cohort the D1cc, D0.03cc, and proportions of the PTV receiving 110%, 115%, and 120% of the prescription dose in the 50 Gy in 4 fraction cohort. Optimal thresholds for these parameters will be identified in further analyses. There did not appear to be an association with LR and BED10, D99%, or comparing single- vs multi-fraction regimens.

Prospective Clinical Adoption of Artificial Intelligence for Organ Contouring in Head and Neck Radiation Treatment Planning

PURPOSE/OBJECTIVE(S)

Patients that undergo head and neck (H&N) radiation therapy (RT) require laborious delineation of organs-at-risk (OARs) on computed tomography (CT) scans in a treatment planning system (TPS) to minimize radiation to normal tissue. This task can be completed rapidly and accurately with recently developed artificial intelligence-based semantic segmentation models. The current study aims to deploy and evaluate a strategy for improving clinical practice with this technology.

MATERIALS/METHODS

Deep learning models were trained and tested with CT scans and OAR contours from previous H&N RT cases at our clinic. Two medical physicists vetted the models and selected a 2.5D U-Net for further implementation. The model was embedded in a dedicated server at the hospital, programmed to read H&N CT scans staged for import into the TPS, generate auto-contours, and write them into a TPS-compatible format made available alongside the scan. In the pilot implementation, the auto-contouring service was utilized for more than 60 cases, prospectively. The auto-contours were quantitatively evaluated against the treatment-approved contours to determine how much modification was performed by the clinical team.

RESULTS

The 2.5D U-Net selected for clinical integration segments 21 OARs in less than 3 minutes per scan. Across all the prospective cases, the mean Dice score and mean 95th percentile Hausdorff distance (mm) between the auto-contour and treatment-approved contour for each of the 21 OARs were as follows, respectively: brainstem (0.93, 1.94), optic chiasm (0.70, 2.96), left cochlea (0.69, 2.37), right cochlea (0.68, 2.44), esophagus (0.88, 2.46), left globe (0.93, 1.50), right globe (0.93, 1.63), glottis (0.91, 2.13), larynx (0.93, 2.76), mandible (0.90, 4.86), left optic nerve (0.78, 1.64), right optic nerve (0.82, 1.65), oral cavity (0.86, 8.46), left parotid gland (0.91, 2.78), right parotid gland (0.91, 2.39), pharynx (0.85, 2.39), spinal cord (0.87, 2.27), left submandibular gland (0.85, 3.46), right submandibular gland (0.83, 3.69), left temporal lobe (0.94, 2.20), and right temporal lobe (0.95, 2.09). The auto-contours for the optic chiasm, optic nerves, cochleas, and submandibular glands differed substantially from the final contours, a finding corroborated by the clinical team; the rest were clinically acceptable with minor or no edits necessary.

CONCLUSION

The proposed strategy provides a sophisticated starting point for treatment planning that has garnered overall favorable feedback from the participating radiation oncologists and dosimetrists. Consequently, the technique is being extended to other treatment sites.

Mitigation of IMRT/SBRT Treatment Planning Errors on the First Biology-Guided Radiotherapy System Using FMEA within Six Sigma Framework

PURPOSE/OBJECTIVE(S)

Utilize the Six Sigma methodology and Failure Mode and Effect Analysis (FMEA) to prevent and mitigate errors in IMRT/SBRT treatment planning with the first clinical installation of biology-guided radiotherapy (BgRT) system.

MATERIALS/METHODS

The Six Sigma approach consisted of five phases: Define-Measure-Analyze-Improve-Control. The Define-Measure-Analyze phases consisted of process mapping and an FMEA of the IMRT/SBRT treatment planning process on the BgRT system. The multidisciplinary team outlined the workflow process and identified the failure modes associated with the plan check items using AAPM TG-100 recommendations. Items with the highest average risk priority numbers (RPN) and Severity ≥7 were prioritized for automation using the treatment planning system scripting API (ESAPI). The Improve phase consisted of developing ESAPI scripts prior to the launch of the BgRT system to improve efficiency and safety. In the Control phase, the FMEA ranking was re-evaluated 1-year post-clinical launch.

RESULTS

Overall, 100 plan check items were identified where the RPN values ranged from 10.2 to 429.0. Fifty of these items (50%) were suitable for automation within ESAPI. Of the 10 highest-risk items (Table 1), 8 were suitable for automation. Based on the results of the FMEA, two scripts were developed: Planning Assistant used by the planner during preparation for planning and the Automated Plan Check used by the planner and the plan checker during plan preparation for treatment. At 1-year post-clinical launch, the scripts were used for 80 patients successfully treated in 1747 fractions. During this period only 3 errors were reported: omitted bolus during treatment, nomenclature error in the BgRT system plan prescription, and dose tracking plan not approved following physics plan check. The average RPN pre-scripts was 138.0 compared to the average post-scripts RPN of 47.8 (p < 0.05) signifying a safer process.

CONCLUSION

Implementing new technology into the clinic can be an error-prone process where the likelihood of errors increases with increasing pressure to implement the technology quickly. To limit errors in the clinical implementation of the first BgRT system, the Six Sigma methodology was utilized to identify failure modes, establish quality control checks, and re-evaluate these checks 1-year post-clinical launch.

Unilateral Diaphragmatic Paralysis After Stereotactic Ablative Radiation Therapy to a Lung Tumor Abutting the Course of the Phrenic Nerve

We present the case of a woman with metastatic adenoid cystic carcinoma who received stereotactic ablative radiation therapy with a total dose of 50 Gy in 4 fractions to 2 lung metastases and developed symptomatic left phrenic nerve injury 2 years after radiation. The maximum dose to the approximate location of the phrenic nerve was 57.7 Gy, which corresponds to a biologically effective dose for late effects (using α/β ratio = 3) of 335.14 Gy. Here, we discuss the case, planning considerations by radiation oncologists and medical physicists, and the multidisciplinary medical management of this patient.

Cancer Informatics for Cancer Centers: Sharing Ideas on How to Build an Artificial Intelligence-Ready Informatics Ecosystem for Radiation Oncology

In August 2022, the Cancer Informatics for Cancer Centers brought together cancer informatics leaders for its biannual symposium, Precision Medicine Applications in Radiation Oncology, co-chaired by Quynh-Thu Le, MD (Stanford University), and Walter J. Curran, MD (GenesisCare). Over the course of 3 days, presenters discussed a range of topics relevant to radiation oncology and the cancer informatics community more broadly, including biomarker development, decision support algorithms, novel imaging tools, theranostics, and artificial intelligence (AI) for the radiotherapy workflow. Since the symposium, there has been an impressive shift in the promise and potential for integration of AI in clinical care, accelerated in large part by major advances in generative AI. AI is now poised more than ever to revolutionize cancer care. Radiation oncology is a field that uses and generates a large amount of digital data and is therefore likely to be one of the first fields to be transformed by AI. As experts in the collection, management, and analysis of these data, the informatics community will take a leading role in ensuring that radiation oncology is prepared to take full advantage of these technological advances. In this report, we provide highlights from the symposium, which took place in Santa Barbara, California, from August 29 to 31, 2022. We discuss lessons learned from the symposium for data acquisition, management, representation, and sharing, and put these themes into context to prepare radiation oncology for the successful and safe integration of AI and informatics technologies.

Pulmonary Hemorrhage in Patients Treated With Thoracic Stereotactic Ablative Radiotherapy and Antiangiogenic Agents

INTRODUCTION

Severe pulmonary hemorrhage can occur in patients treated with thoracic stereotactic ablative radiotherapy (SABR) and vascular endothelial growth factor inhibitors (VEGFis). There is limited understanding of which patients are at risk for toxicity with the combination of thoracic SABR and VEGFis or how the risk differs over either therapy alone.

METHODS

We evaluated a prospectively maintained cohort of 690 patients with 818 pulmonary tumors treated with highly conformal SABR. Rates of any-grade and grade 3 plus (G3+) pulmonary hemorrhage were compared between patients treated with or without VEGFi therapy across tumor locations. Outcomes were compared between patients treated with SABR plus VEGFi and a propensity-matched cohort of those treated with VEGFi therapy alone.

RESULTS

Treatment with VEGFi plus SABR was associated with higher rates of G3+ pulmonary hemorrhage compared with those treated with SABR alone for the overall cohort (3-y incidence: 7.9% versus 0.6%, p < 0.01) and those with central tumors (19.1% versus 3.3%, p = 0.04). When further subdivided, there were significantly higher toxicity rates with VEGFi for the ultracentral (9.0% versus 45.0%, p = 0.044), but not central nonabutting tumors (0.0% versus 1.3%, p = 0.69). There was an increased incidence of G3+ hemorrhage in patients treated with VEGFi plus SABR compared with VEGFi alone (9.6% versus 1.3%, p = 0.04).

CONCLUSIONS

The combination of VEGFi and SABR was associated with an increased risk of high-grade pulmonary hemorrhage over either therapy alone. Low rates of toxicity were observed when excluding patients with SABR to ultracentral tumors and applying highly conformal SABR techniques.

Study of Patient and Physician Attitudes Toward Automated Prognostic Models for Patients With Metastatic Cancer

PURPOSE

For patients with cancer and their doctors, prognosis is important for choosing treatments and supportive care. Oncologists' life expectancy estimates are often inaccurate, and many patients are not aware of their general prognosis. Machine learning (ML) survival models could be useful in the clinic, but there are potential concerns involving accuracy, provider training, and patient involvement. We conducted a qualitative study to learn about patient and oncologist views on potentially using a ML model for patient care.

METHODS

Patients with metastatic cancer (n = 15) and their family members (n = 5), radiation oncologists (n = 5), and medical oncologists (n = 5) were recruited from a single academic health system. Participants were shown an anonymized report from a validated ML survival model for another patient, which included a predicted survival curve and a list of variables influencing predicted survival. Semistructured interviews were conducted using a script.

RESULTS

Every physician and patient who completed their interview said that they would want the option for the model to be used in their practice or care. Physicians stated that they would use an AI prognosis model for patient triage and increasing patient understanding, but had concerns about accuracy and explainability. Patients generally said that they would trust model results completely if presented by their physician but wanted to know if the model was being used in their care. Some reacted negatively to being shown a median survival prediction.

CONCLUSION

Patients and physicians were supportive of use of the model in the clinic, but had various concerns, which should be addressed as predictive models are increasingly deployed in practice.

Lessons and Opportunities for Biomarker-Driven Radiation Personalization in Head and Neck Cancer

Head and neck cancer is notoriously challenging to treat in part because it constitutes an anatomically and biologically diverse group of cancers with heterogeneous prognoses. While treatment can be associated with significant late toxicities, recurrence is often difficult to salvage with poor survival rates and functional morbidity.^1,2 Thus, achieving tumor control and cure at the initial diagnosis is the highest priority. Given the differing outcome expectations (even within a specific sub-site like oropharyngeal carcinoma), there has been growing interest in personalizing treatment: de-escalation in selected cancers to decrease the risk of late toxicity without compromising oncologic outcomes, and intensification for more aggressive cancers to improve oncologic outcomes without causing undue toxicity. This risk stratification is increasingly accomplished using biomarkers, which can represent molecular, clinicopathologic, and/or radiologic data. In this review, we will focus on biomarker-driven radiotherapy dose personalization with emphasis on oropharyngeal and nasopharyngeal carcinoma. This radiation personalization is largely performed on the population level by identifying patients with good prognosis via traditional clinicopathologic factors, although there are emerging studies supporting inter-tumor and intra-tumor level personalization via imaging and molecular biomarkers.

Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study

BACKGROUND

Only around 20-30% of patients with non-small-cell lung cancer (NCSLC) have durable benefit from immune-checkpoint inhibitors. Although tissue-based biomarkers (eg, PD-L1) are limited by suboptimal performance, tissue availability, and tumour heterogeneity, radiographic images might holistically capture the underlying cancer biology. We aimed to investigate the application of deep learning on chest CT scans to derive an imaging signature of response to immune checkpoint inhibitors and evaluate its added value in the clinical context.

METHODS

In this retrospective modelling study, 976 patients with metastatic, EGFR/ALK negative NSCLC treated with immune checkpoint inhibitors at MD Anderson and Stanford were enrolled from Jan 1, 2014, to Feb 29, 2020. We built and tested an ensemble deep learning model on pretreatment CTs (Deep-CT) to predict overall survival and progression-free survival after treatment with immune checkpoint inhibitors. We also evaluated the added predictive value of the Deep-CT model in the context of existing clinicopathological and radiological metrics.

FINDINGS

Our Deep-CT model demonstrated robust stratification of patient survival of the MD Anderson testing set, which was validated in the external Stanford set. The performance of the Deep-CT model remained significant on subgroup analyses stratified by PD-L1, histology, age, sex, and race. In univariate analysis, Deep-CT outperformed the conventional risk factors, including histology, smoking status, and PD-L1 expression, and remained an independent predictor after multivariate adjustment. Integrating the Deep-CT model with conventional risk factors demonstrated significantly improved prediction performance, with overall survival C-index increases from 0·70 (clinical model) to 0·75 (composite model) during testing. On the other hand, the deep learning risk scores correlated with some radiomics features, but radiomics alone could not reach the performance level of deep learning, indicating that the deep learning model effectively captured additional imaging patterns beyond known radiomics features.

INTERPRETATION

This proof-of-concept study shows that automated profiling of radiographic scans through deep learning can provide orthogonal information independent of existing clinicopathological biomarkers, bringing the goal of precision immunotherapy for patients with NSCLC closer.

FUNDING

National Institutes of Health, Mark Foundation Damon Runyon Foundation Physician Scientist Award, MD Anderson Strategic Initiative Development Program, MD Anderson Lung Moon Shot Program, Andrea Mugnaini, and Edward L C Smith.

Stratified assessment of an FDA-cleared deep learning algorithm for automated detection and contouring of metastatic brain tumors in stereotactic radiosurgery

PURPOSE

Artificial intelligence-based tools can be leveraged to improve detection and segmentation of brain metastases for stereotactic radiosurgery (SRS). VBrain by Vysioneer Inc. is a deep learning algorithm with recent FDA clearance to assist in brain tumor contouring. We aimed to assess the performance of this tool by various demographic and clinical characteristics among patients with brain metastases treated with SRS.

MATERIALS AND METHODS

We randomly selected 100 patients with brain metastases who underwent initial SRS on the CyberKnife from 2017 to 2020 at a single institution. Cases with resection cavities were excluded from the analysis. Computed tomography (CT) and axial T1-weighted post-contrast magnetic resonance (MR) image data were extracted for each patient and uploaded to VBrain. A brain metastasis was considered "detected" when the VBrain- "predicted" contours overlapped with the corresponding physician contours ("ground-truth" contours). We evaluated performance of VBrain against ground-truth contours using the following metrics: lesion-wise Dice similarity coefficient (DSC), lesion-wise average Hausdorff distance (AVD), false positive count (FP), and lesion-wise sensitivity (%). Kruskal-Wallis tests were performed to assess the relationships between patient characteristics including sex, race, primary histology, age, and size and number of brain metastases, and performance metrics such as DSC, AVD, FP, and sensitivity.

RESULTS

We analyzed 100 patients with 435 intact brain metastases treated with SRS. Our cohort consisted of patients with a median number of 2 brain metastases (range: 1 to 52), median age of 69 (range: 19 to 91), and 50% male and 50% female patients. The primary site breakdown was 56% lung, 10% melanoma, 9% breast, 8% gynecological, 5% renal, 4% gastrointestinal, 2% sarcoma, and 6% other, while the race breakdown was 60% White, 18% Asian, 3% Black/African American, 2% Native Hawaiian or other Pacific Islander, and 17% other/unknown/not reported. The median tumor size was 0.112 c.c. (range: 0.010-26.475 c.c.). We found mean lesion-wise DSC to be 0.723, mean lesion-wise AVD to be 7.34% of lesion size (0.704 mm), mean FP count to be 0.72 tumors per case, and lesion-wise sensitivity to be 89.30% for all lesions. Moreover, mean sensitivity was found to be 99.07%, 97.59%, and 96.23% for lesions with diameter equal to and greater than 10 mm, 7.5 mm, and 5 mm, respectively. No other significant differences in performance metrics were observed across demographic or clinical characteristic groups.

CONCLUSION

In this study, a commercial deep learning algorithm showed promising results in segmenting brain metastases, with 96.23% sensitivity for metastases with diameters of 5 mm or higher. As the software is an assistive AI, future work of VBrain integration into the clinical workflow can provide further clinical and research insights.

Safety of Nivolumab Added to Chemoradiation Therapy Platforms for Intermediate and High-Risk Locoregionally Advanced Head and Neck Squamous Cell Carcinoma: RTOG Foundation 3504

PURPOSE

Programmed death-1 immune checkpoint blockade improves survival of patients with recurrent/metastatic head and neck squamous cell carcinoma (HNSCC), but the benefits of addition to (chemo)radiation for newly diagnosed patients with HNSCC remain unknown.

METHODS AND MATERIALS

We evaluated the safety of nivolumab concomitant with 70 Gy intensity modulated radiation therapy and weekly cisplatin (arm 1), every 3-week cisplatin (arm 2), cetuximab (arm 3), or alone for platinum-ineligible patients (arm 4) in newly diagnosed intermediate- or high-risk locoregionally advanced HNSCC. Patients received nivolumab from 2 weeks prior to radiation therapy until 3 months post-radiation therapy. The primary endpoint was dose-limiting toxicity (DLT). If ≤2 of the first 8 evaluable patients experienced a DLT, an arm was considered safe. Secondary endpoints included toxicity and feasibility of adjuvant nivolumab to 1 year, defined as all 7 additional doses received by ≥4 of the first 8 evaluable patients across arms.

RESULTS

Of 39 patients (10 in arms 1, 3, 4 and 9 in arm 2), 72% had T3-4 tumors, 85% had N2-3 nodal disease, and 67% had >10 pack-years of smoking. There were no DLTs in arms 1 and 2, 1 in arm 3 (mucositis), and 2 in arm 4 (lipase elevation and mucositis in 1 and fatigue in another). The most common grade ≥3 nivolumab-related adverse events were lipase increase, mucositis, diarrhea, lymphopenia, hyponatremia, leukopenia, fatigue, and serum amylase increase. Adjuvant nivolumab was feasible as defined in the protocol.

CONCLUSIONS

Concomitant nivolumab with the 4 tested regimens was safe for patients with intermediate- and high-risk HNSCC, and subsequent adjuvant nivolumab was feasible as defined (NCT02764593).

Use of Machine Learning and Lay Care Coaches to Increase Advance Care Planning Conversations for Patients With Metastatic Cancer

PURPOSE

Patients with metastatic cancer benefit from advance care planning (ACP) conversations. We aimed to improve ACP using a computer model to select high-risk patients, with shorter predicted survival, for conversations with providers and lay care coaches. Outcomes included ACP documentation frequency and end-of-life quality measures.

METHODS

In this study of a quality improvement initiative, providers in four medical oncology clinics received Serious Illness Care Program training. Two clinics (thoracic/genitourinary) participated in an intervention, and two (cutaneous/sarcoma) served as controls. ACP conversations were documented in a centralized form in the electronic medical record. In the intervention, providers and care coaches received weekly e-mails highlighting upcoming clinic patients with < 2 year computer-predicted survival and no prior prognosis documentation. Care coaches contacted these patients for an ACP conversation (excluding prognosis). Providers were asked to discuss and document prognosis.

RESULTS

In the four clinics, 4,968 clinic visits by 1,251 patients met inclusion criteria (metastatic cancer with no prognosis previously documented). In their first visit, 28% of patients were high-risk (< 2 year predicted survival). Preintervention, 3% of both intervention and control clinic patients had ACP documentation during a visit. By intervention end (February 2021), 35% of intervention clinic patients had ACP documentation compared with 3% of control clinic patients. Providers' prognosis documentation rate also increased in intervention clinics after the intervention (2%-27% in intervention clinics, P < .0001; 0%-1% in control clinics). End-of-life care intensity was similar in intervention versus control clinics, but patients with ≥ 1 provider ACP edit met fewer high-intensity care measures (P = .04).

CONCLUSION

Combining a computer prognosis model with care coaches increased ACP documentation.

Improving lung cancer screening rates among patients with head and neck cancer in a radiation oncology clinic

Background

The United States Preventive Services Task Force (USPSTF) recommends lung cancer screening via annual low dose computed tomography (LDCT) for high risk patients. Despite the strong evidence of a mortality benefit from several randomized clinical trials, rates of lung cancer screening remain low. We plan to assess how screening guidelines are implemented in a radiation oncology clinic for patients with head and neck cancer.

Methods

A single institution, retrospective chart review was used to identify patients with head and neck cancer seen in a radiation oncology clinic who were potentially eligible for lung cancer screening under the current USPSTF guidelines. Patients who were potentially screening-eligible were enrolled in a phone survey to assess their knowledge about lung cancer screening and willingness to be screened.

Results

Of the 184 patients with head and neck cancer seen in the clinic, 8 (4%) patients were eligible for lung cancer screening under the previous USPSTF recommendations, including 1 (0.5%) patient already being screened. One patient (0.5%) became eligible under the expanded guidelines. All 184 patients had smoking history documented. Of the 87 current or former smokers, there were 24 (28%) who did not have pack-years documented; of the 82 former smokers, there were 8 (10%) who did not have quit date documented. Among the 16 phone survey participants (response rate: 70%) only 6 (38%) were aware there is a way to screen for lung cancer and 12 (75%) patients would be interested in screening if they are found to be eligible.

Conclusions

These findings highlight a potential opportunity to increase rates of lung cancer screening among patients with head and neck cancer by both enhancing provider awareness as well as patient education at the community level.

Considerations in the reliability and fairness audits of predictive models for advance care planning

Multiple reporting guidelines for artificial intelligence (AI) models in healthcare recommend that models be audited for reliability and fairness. However, there is a gap of operational guidance for performing reliability and fairness audits in practice. Following guideline recommendations, we conducted a reliability audit of two models based on model performance and calibration as well as a fairness audit based on summary statistics, subgroup performance and subgroup calibration. We assessed the Epic End-of-Life (EOL) Index model and an internally developed Stanford Hospital Medicine (HM) Advance Care Planning (ACP) model in 3 practice settings: Primary Care, Inpatient Oncology and Hospital Medicine, using clinicians' answers to the surprise question (“Would you be surprised if [patient X] passed away in [Y years]?”) as a surrogate outcome. For performance, the models had positive predictive value (PPV) at or above 0.76 in all settings. In Hospital Medicine and Inpatient Oncology, the Stanford HM ACP model had higher sensitivity (0.69, 0.89 respectively) than the EOL model (0.20, 0.27), and better calibration (O/E 1.5, 1.7) than the EOL model (O/E 2.5, 3.0). The Epic EOL model flagged fewer patients (11%, 21% respectively) than the Stanford HM ACP model (38%, 75%). There were no differences in performance and calibration by sex. Both models had lower sensitivity in Hispanic/Latino male patients with Race listed as “Other.” 10 clinicians were surveyed after a presentation summarizing the audit. 10/10 reported that summary statistics, overall performance, and subgroup performance would affect their decision to use the model to guide care; 9/10 said the same for overall and subgroup calibration. The most commonly identified barriers for routinely conducting such reliability and fairness audits were poor demographic data quality and lack of data access. This audit required 115 person-hours across 8–10 months. Our recommendations for performing reliability and fairness audits include verifying data validity, analyzing model performance on intersectional subgroups, and collecting clinician-patient linkages as necessary for label generation by clinicians. Those responsible for AI models should require such audits before model deployment and mediate between model auditors and impacted stakeholders.

Posttreatment FDG-PET/CT Hopkins criteria predict locoregional recurrence after definitive radiotherapy for oropharyngeal squamous cell carcinoma

BACKGROUND

Metabolic response assessment for oropharyngeal squamous cell carcinoma (OPSCC) aids in identifying locoregional persistence/recurrence (LRR). The Hopkins Criteria are a standardized qualitative response assessment system using posttreatment FDG-PET/CT.

METHODS

We conducted a retrospective cohort study of patients with node-positive OPSCC treated with definitive (chemo)radiotherapy. We assessed Hopkins Criteria performance for LRR, then developed and validated a competing-risks model.

RESULTS

Between 2004 and 2018, 259 patients were included with median follow-up of 43 months. The Hopkins Criteria sensitivity, specificity, negative predictive value, and accuracy were 68%, 88%, 95%, and 85%. The 36-month cumulative incidence of LRR was greater with positive scores (45% vs. 5%, HR 12.60, p < 0.001). PET/CTs performed ≤10 weeks after radiotherapy were associated with a four-fold increase in pathologically negative biopsies/surgeries (36% vs. 9%, p = 0.03). The AUC for LRR was 0.89 using a model integrating the Hopkins score.

CONCLUSIONS

The Hopkins Criteria predict LRR with high accuracy for OPSCC response assessment.

Use of systemic cancer treatments based on a validated survival prediction model in metastatic cancer

e13515

Background: Use of systemic anti-cancer treatment near the end of life (EOL) is recognized as a low value practice with limited benefit to patients. Machine learning (ML) models that identify patients in close proximity to death can help prospectively assess oncology practice of systemic therapy use. We hypothesized that systemic therapy use would be higher based on predicted survival compared with actual survival. Methods: We calculated prevalence of systemic therapy use based on predicted and actual survival among patients with metastatic cancer at Stanford Healthcare from 2008-2019. Patients were included if they were in the test set of the ML model, had an eligible outpatient oncology clinic visit for which a predicted survival was calculated and were deceased . Median predicted survival was calculated from the ML model at each outpatient oncology visit and treatment was linked to a visit date if within 14 days of each other. Prevalence of systemic therapy was calculated for patients with a predicted or actual survival of < 6 months, 6-12 months, 12-18 months and 18-24 months. The five categories of treatment were: chemotherapy, targeted/antibody, hormone, immunotherapy, and other. Results: A total of 951 deceased patients who received anticancer treatment are included and a total of 21,283 doses of treatment were administered with a mean of 22 doses per patient. The median age at metastatic cancer diagnosis was 58 years, 53% of patients were female and most patients identified as White (55%) or Asian (23%). The most common disease groups were gastrointestinal (21.6%), thoracic (18.6%) and breast (14.9%). Overall, the use of different treatment types did not differ based on either predicted or actual survival (Table). In all the survival groupings, chemotherapy remained the predominant medication type, however with a trend of decreasing use with longer predicted and actual survival. Conclusions: The use of cancer medications and the type of medication given did not change based on predicted or actual survival in a large group of patients with metastatic cancer. There was a trend of decreasing chemotherapy use with longer prognosis. Further investigation into use in time intervals closer to (predicted or actual) death and inclusion of those who did not receive any systemic therapy are underway.[Table: see text]

Accuracy of Electronic Medical Record Follow-Up Data for Estimating the Survival Time of Patients With Cancer

PURPOSE

For real-world evidence, it is convenient to use routinely collected data from the electronic medical record (EMR) to measure survival outcomes. However, patients can become lost to follow-up, causing incomplete data and biased survival time estimates. We quantified this issue for patients with metastatic cancer seen in an academic health system by comparing survival estimates from EMR data only and from EMR data combined with high-quality cancer registry data.

MATERIALS AND METHODS

Patients diagnosed with metastatic cancer from 2008 to 2014 were included in this retrospective study. Patients who were diagnosed with cancer or received their initial treatment within our system were included in the institutional cancer registry and this study. Overall survival was calculated using the Kaplan-Meier method. Survival curves were generated in two ways: using EMR follow-up data alone and using EMR data supplemented with data from the Stanford Cancer Registry/California Cancer Registry.

RESULTS

Four thousand seventy-seven patients were included. The median follow-up using EMR + Cancer Registry data was 19.9 months, and the median follow-up in surviving patients was 67.6 months. There were 1,301 deaths recorded in the EMR and 3,140 deaths recorded in the Cancer Registry. The median overall survival from the date of cancer diagnosis using EMR data was 58.7 months (95% CI, 54.2 to 63.2); using EMR + Cancer Registry data, it was 20.8 months (95% CI, 19.6 to 22.3). A similar pattern was seen using the date of first systemic therapy or date of first hospital admission as the baseline date.

CONCLUSION

Using EMR data alone, survival time was overestimated compared with EMR + Cancer Registry data.

Uncovering interpretable potential confounders in electronic medical records

Randomized clinical trials (RCT) are the gold standard for informing treatment decisions. Observational studies are often plagued by selection bias, and expert-selected covariates may insufficiently adjust for confounding. We explore how unstructured clinical text can be used to reduce selection bias and improve medical practice. We develop a framework based on natural language processing to uncover interpretable potential confounders from text. We validate our method by comparing the estimated hazard ratio (HR) with and without the confounders against established RCTs. We apply our method to four cohorts built from localized prostate and lung cancer datasets from the Stanford Cancer Institute and show that our method shifts the HR estimate towards the RCT results. The uncovered terms can also be interpreted by oncologists for clinical insights. We present this proof-of-concept study to enable more credible causal inference using observational data, uncover meaningful insights from clinical text, and inform high-stakes medical decisions.

Randomized clinical trials are often plagued by selection bias, and expert-selected covariates may insufficiently adjust for confounding factors. Here, the authors develop a framework based on natural language processing to uncover interpretable potential confounders from text.

Acute and Late Esophageal Toxicity Following Stereotactic Ablative Radiotherapy to Thoracic Tumors near or Abutting the Esophagus

PURPOSE

To evaluate the incidence of acute and late esophageal toxicity in patients with thoracic tumors near or abutting the esophagus treated with stereotactic ablative radiotherapy (SABR).

METHODS AND MATERIALS

Among patients with thoracic tumors treated with SABR, we identified those with tumors near or abutting the esophagus. Using the linear-quadratic model with an α/ß ratio of 10, we determined the correlation between dosimetric parameters and esophageal toxicity graded using the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0.

RESULTS

Out of 2200 patients treated with thoracic SABR, 767 patients were analyzable for esophageal dosimetry. We identified 55 patients with tumors near the esophagus (52 evaluable for esophagitis grade), 28 with PTV overlapping the esophagus. Median follow-up and overall survival were 16 and 23 months respectively. Thirteen patients (25%) developed temporary grade 2 acute esophageal toxicity, 11 (85%) of whom had PTV overlapping the esophagus. Symptoms resolved within 1-3 months in 12 patients, and 6 months in all patients. No grade 3-5 toxicity was observed. Only 3 patients (6%) developed late or persistent grade 2 dysphagia or dyspepsia of uncertain relationship to SABR. Cumulative incidence of acute esophagitis was 15% and 25% at 14 days and 60 days respectively. Acute toxicity correlated on univariate analysis with esophageal D_max, D_1cc, D_2cc, D_max/D_prescription and whether the PTV was overlapping the esophagus. Esophageal D_max (BED₁₀) < 62 Gy, D_1cc (BED₁₀) < 48 Gy, D_2cc (BED₁₀) < 43 Gy, and D_max/D_prescription < 85% was associated with <20% risk of grade 2 acute esophagitis. Only 2 local recurrences occurred.

CONCLUSIONS

Although 25% of patients with tumors near the esophagus developed acute esophagitis (39% of those with PTV overlapping the esophagus), these toxicities were all grade 2 and all temporary. This suggests the safety and efficacy of thoracic SABR for tumors near or abutting the esophagus when treating with high conformity and sharp dose gradients.

Use of a computer model and care coaches to increase advance care planning conversations for patients with metastatic cancer

8

Background: Patients with metastatic cancer benefit from advance care planning (ACP) conversations. Despite initiatives which train providers to have ACP conversations using the serious illness care program (SICP) conversation guide, few patients have a documented prognosis discussion due to busy clinic schedules and difficulty in deciding the right times to have such conversations. We designed an intervention to improve ACP by incorporating a validated computer model to identify patients at high risk for mortality in combination with lay care coaches. We investigated whether this would improve end of life quality measures. Methods: Four Stanford clinics were included in this pilot; all received SICP training. Two clinics (thoracic and genitourinary) underwent the intervention (computer model + care coach), and two clinics (sarcoma and cutaneous) served as the control. For providers in the intervention, an email was sent every Sunday listing the metastatic cancer patients who would be seen in clinic the following week and a predicted prognosis generated by the model. A lay care coach contacted patients with a predicted survival ≤2 years to have an ACP conversation with them. After, the care coach notified the provider to suggest discussion regarding prognosis with the patient. Criteria for a patient visit to be included in the analysis were: age ≥18, established patient, has sufficient EMR data for computer model, and no prior prognosis documentation. The primary outcome was documentation of prognosis in the ACP form by the end of the week following the clinic visit. Results: 5330 visits in 1298 unique patients met the inclusion criteria. Median age was 67 (range 19-97); 790 male, 508 female. 1970 visits were with patients with ≤2 year predicted survival. Prognosis discussion was documented by providers in the ACP form for 8.1% of intervention visits compared to 0.07% of control visits (p=0.001 in mixed effects model). Of the 1298 unique patients, 84 were deceased by December 2020. 41.7% died in the hospital. 59.5% were enrolled in hospice prior to death, and 19.0% were hospitalized in the ICU ≤14 days prior to death. Of deceased patients with ACP form prognosis documentation, 5.0% had ≥2 hospitalizations in the 30 days before death compared to 23.4% of deceased patients with no prognosis documented (p=0.10). For ≥ 2 ER visits in the 30 days before death, the proportions were 5.0% and 20.3% (p=0.17). Conclusions: This pilot study supports that our intervention is associated with higher rates of prognosis discussions and documentation. There was a trend towards better quality of end of life care as noted by higher rates of hospice enrollment and less intensive care at end of life. These results merit further investigation as a means to improve goal-concordant care and ensure appropriate care for cancer patients at the end of life.

Natural Language Processing to Identify Cancer Treatments With Electronic Medical Records

PURPOSE

Knowing the treatments administered to patients with cancer is important for treatment planning and correlating treatment patterns with outcomes for personalized medicine study. However, existing methods to identify treatments are often lacking. We develop a natural language processing approach with structured electronic medical records and unstructured clinical notes to identify the initial treatment administered to patients with cancer.

METHODS

We used a total number of 4,412 patients with 483,782 clinical notes from the Stanford Cancer Institute Research Database containing patients with nonmetastatic prostate, oropharynx, and esophagus cancer. We trained treatment identification models for each cancer type separately and compared performance of using only structured, only unstructured (bag-of-words, doc2vec, fasttext), and combinations of both (structured + bow, structured + doc2vec, structured + fasttext). We optimized the identification model among five machine learning methods (logistic regression, multilayer perceptrons, random forest, support vector machines, and stochastic gradient boosting). The treatment information recorded in the cancer registry is the gold standard and compares our methods to an identification baseline with billing codes.

RESULTS

For prostate cancer, we achieved an f1-score of 0.99 (95% CI, 0.97 to 1.00) for radiation and 1.00 (95% CI, 0.99 to 1.00) for surgery using structured + doc2vec. For oropharynx cancer, we achieved an f1-score of 0.78 (95% CI, 0.58 to 0.93) for chemoradiation and 0.83 (95% CI, 0.69 to 0.95) for surgery using doc2vec. For esophagus cancer, we achieved an f1-score of 1.0 (95% CI, 1.0 to 1.0) for both chemoradiation and surgery using all combinations of structured and unstructured data. We found that employing the free-text clinical notes outperforms using the billing codes or only structured data for all three cancer types.

CONCLUSION

Our results show that treatment identification using free-text clinical notes greatly improves upon the performance using billing codes and simple structured data. The approach can be used for treatment cohort identification and adapted for longitudinal cancer treatment identification.

Automated model versus treating physician for predicting survival time of patients with metastatic cancer

OBJECTIVE

Being able to predict a patient's life expectancy can help doctors and patients prioritize treatments and supportive care. For predicting life expectancy, physicians have been shown to outperform traditional models that use only a few predictor variables. It is possible that a machine learning model that uses many predictor variables and diverse data sources from the electronic medical record can improve on physicians' performance. For patients with metastatic cancer, we compared accuracy of life expectancy predictions by the treating physician, a machine learning model, and a traditional model.

MATERIALS AND METHODS

A machine learning model was trained using 14 600 metastatic cancer patients' data to predict each patient's distribution of survival time. Data sources included note text, laboratory values, and vital signs. From 2015-2016, 899 patients receiving radiotherapy for metastatic cancer were enrolled in a study in which their radiation oncologist estimated life expectancy. Survival predictions were also made by the machine learning model and a traditional model using only performance status. Performance was assessed with area under the curve for 1-year survival and calibration plots.

RESULTS

The radiotherapy study included 1190 treatment courses in 899 patients. A total of 879 treatment courses in 685 patients were included in this analysis. Median overall survival was 11.7 months. Physicians, machine learning model, and traditional model had area under the curve for 1-year survival of 0.72 (95% CI 0.63-0.81), 0.77 (0.73-0.81), and 0.68 (0.65-0.71), respectively.

CONCLUSIONS

The machine learning model's predictions were more accurate than those of the treating physician or a traditional model.

Postoperative Observation Versus Radiotherapy for Pathologic N1 Oral Cavity Squamous Cell Carcinoma

OBJECTIVES

To investigate the benefit of postoperative radiotherapy (PORT) for low-volume (pN1) nodal disease after resection of oral cavity squamous cell carcinoma.

MATERIALS AND METHODS

The National Cancer Database was queried for adults with nonmetastatic squamous cell carcinoma of the oral cavity treated by surgical resection with pathologic stage T1-2 N0-2 (American Joint Committee on Cancer 7th edition) and with the maximal exclusion of standard indications for PORT. Overall survival was compared within pN1 for observation versus PORT and then compared for pN1 versus pN0 and versus pN2 stratified by receipt of observation or PORT. Multivariable Cox regression was used to adjust for potential confounders between PORT and survival, including comorbidity and age.

RESULTS

Overall 5017 pN0, 530 pN1, and 253 pN2 patients were identified, of whom 9%, 35%, and 64% received PORT, respectively. Within the pN1 cohort, PORT was associated with improved survival versus observation (adjusted hazard ratio, 0.66; 95% confidence interval, 0.46-0.97; P=0.03). Among observed patients, the prognosis of pN1 was equivalent to pN2 and inferior to pN0; in contrast, among patients treated with PORT, the prognosis of pN1 was equivalent to pN0 and superior to pN2. Without PORT, pN1 remained an adverse risk factor relative to pN0 regardless of the depth of invasion, lymph node size, lymph node location, and extent of lymph node dissection.

CONCLUSIONS

PORT was associated with a survival benefit compared with observation. Notably, pN1 was an adverse risk factor relative to pN0 if, and only if, patients did not receive PORT, suggesting pN1 by itself may be an indication for PORT.

Prolongation of definitive head and neck cancer radiotherapy: Survival impact and predisposing factors

BACKGROUND AND PURPOSE

To quantify the survival impact of prolongation of definitive radiotherapy (RT) for head and neck cancer in a national, modern cohort, and to identify predictive factors for prolongation.

MATERIALS AND METHODS

The National Cancer Database was queried for adults with non-metastatic cancer of the nasopharynx, oropharynx, larynx, or hypopharynx diagnosed 2004-2015, treated with definitive RT to 66-70 Gy in 30-35 fractions at 2-2.2 Gy per fraction. Multivariable Cox regression and propensity score matching were used to model the survival impact of RT prolongation, adjusting for potential confounders such as age and comorbidity. Predictors of RT prolongation were identified using multivariable multinomial logistic regression.

RESULTS

In total, 36,367 patients were identified. As a continuous variable, RT prolongation increased the relative hazard of death by 2% per day (P < .0001). In the matched cohorts, patients with short (4-8 days) or long prolongation (>8 days) had lower absolute 4-year overall survival by 4% and 12%, respectively (P < .0001), while prolongation of 1-3 days was not significantly adverse. Major predictors of increased risk of prolongation were administration of systemic therapy, baseline comorbidity, lack of private insurance, and tumor/nodal stage. Conversely, higher facility volume was significantly protective, with a 55% lower risk of long prolongation within the topmost quartile (>11.5 patients/year).

CONCLUSION

RT prolongation, especially >8 days, is significantly deleterious. Systemic therapy and facility volume were major predictors. Early identification of patients at increased risk of treatment interruptions may facilitate implementation of preventive measures.

Early response evaluation using primary tumor and nodal imaging features to predict progression-free survival of locally advanced non-small cell lung cancer

Prognostic biomarkers that can reliably predict early disease progression of non-small cell lung cancer (NSCLC) are needed for identifying those patients at high risk for progression, who may benefit from more intensive treatment. In this work, we aimed to identify an imaging signature for predicting progression-free survival (PFS) of locally advanced NSCLC. Methods: This retrospective study included 82 patients with stage III NSCLC treated with definitive chemoradiotherapy for whom both baseline and mid-treatment PET/CT scans were performed. They were randomly placed into two groups: training cohort (n=41) and testing cohort (n=41). All primary tumors and involved lymph nodes were delineated. Forty-five quantitative imaging features were extracted to characterize the tumors and involved nodes at baseline and mid-treatment as well as differences between two scans performed at these two points. An imaging signature was developed to predict PFS by fitting an L1-regularized Cox regression model. Results: The final imaging signature consisted of three imaging features: the baseline tumor volume, the baseline maximum distance between involved nodes, and the change in maximum distance between the primary tumor and involved nodes measured at two time points. According to multivariate analysis, the imaging model was an independent prognostic factor for PFS in both the training (hazard ratio [HR], 1.14, 95% confidence interval [CI], 1.04-1.24; P = 0.003), and testing (HR, 1.21, 95% CI, 1.10-1.33; P = 0.048) cohorts. The imaging signature stratified patients into low- and high-risk groups, with 2-year PFS rates of 61.9% and 33.2%, respectively (P = 0.004 [log-rank test]; HR, 4.13, 95% CI, 1.42-11.70) in the training cohort, as well as 43.8% and 22.6%, respectively (P = 0.006 [log-rank test]; HR, 3.45, 95% CI, 1.35-8.83) in the testing cohort. In both cohorts, the imaging signature significantly outperformed conventional imaging metrics, including tumor volume and SUV_max value (C-indices: 0.77-0.79 for imaging signature, and 0.53-0.73 for conventional metrics). Conclusions: Evaluation of early treatment response by combining primary tumor and nodal imaging characteristics may improve the prediction of PFS of locally advanced NSCLC patients.

Predicting per-lesion local recurrence in locally advanced non-small cell lung cancer following definitive radiation therapy using pre- and mid-treatment metabolic tumor volume

BACKGROUND

We evaluated whether pre- and mid-treatment metabolic tumor volume (MTV) predicts per lesion local recurrence (LR) in patients treated with definitive radiation therapy (RT, dose≥60 Gy) for locally advanced non-small cell lung cancer (NSCLC).

METHODS

We retrospectively reviewed records of patients with stage III NSCLC treated from 2006 to 2018 with pre- and mid-RT PET-CT. We measured the MTV of treated lesions on the pre-RT (MTV_pre) and mid-RT (MTV_mid) PET-CT. LR was defined per lesion as recurrence within the planning target volume. Receiver operating characteristic (ROC) curves, cumulative incidence rates, and uni- and multivariable (MVA) competing risk regressions were used to evaluate the association between MTV and LR.

RESULTS

We identified 111 patients with 387 lesions (112 lung tumors and 275 lymph nodes). Median age was 68 years, 69.4% were male, 46.8% had adenocarcinoma, 39.6% had squamous cell carcinoma, and 95.5% received concurrent chemotherapy. Median follow-up was 38.7 months. 3-year overall survival was 42.3%. 3-year cumulative incidence of LR was 26.8% per patient and 11.9% per lesion. Both MTV_pre and MTV_mid were predictive of LR by ROC (AUC = 0.71 and 0.76, respectively) and were significantly associated with LR on MVA (P = 0.004 and P = 7.1e-5, respectively). Among lesions at lower risk of LR based on MTV_pre, higher MTV_mid was associated with LR (P = 0.001).

CONCLUSION

Per-lesion, larger MTV_pre and MTV_mid predicted for increased risk of LR. MTV_mid was more highly predictive of LR than MTV_pre and if validated may allow for further discrimination of high-risk lesions at mid-RT informing dose painting strategies.

Tumor Subregion Evolution-Based Imaging Features to Assess Early Response and Predict Prognosis in Oropharyngeal Cancer

The incidence of oropharyngeal squamous cell carcinoma (OPSCC) has been rapidly increasing. Disease stage and smoking history are often used in current clinical trials to select patients for deintensification therapy, but these features lack sufficient accuracy for predicting disease relapse. Our purpose was to develop an imaging signature to assess early response and predict outcomes of OPSCC. Methods: We retrospectively analyzed 162 OPSCC patients treated with concurrent chemoradiotherapy, equally divided into separate training and validation cohorts with similar clinical characteristics. A robust consensus clustering approach was used to spatially partition the primary tumor and involved lymph nodes into subregions (i.e., habitats) based on 18F-FDG PET and contrast CT imaging. We proposed quantitative image features to characterize the temporal volumetric change of the habitats and peritumoral/nodal tissue between baseline and midtreatment. The reproducibility of these features was evaluated. We developed an imaging signature to predict progression-free survival (PFS) by fitting an L1-regularized Cox regression model. Results: We identified 3 phenotypically distinct intratumoral habitats: metabolically active and heterogeneous, enhancing and heterogeneous, and metabolically inactive and homogeneous. The final Cox model consisted of 4 habitat evolution-based features. In both cohorts, this imaging signature significantly outperformed traditional imaging metrics, including midtreatment metabolic tumor volume for predicting PFS, with a C-index of 0.72 versus 0.67 (training) and 0.66 versus 0.56 (validation). The imaging signature stratified patients into high-risk versus low-risk groups with 2-y PFS rates of 59.1% versus 89.4% (hazard ratio, 4.4; 95% confidence interval, 1.4-13.4 [training]) and 61.4% versus 87.8% (hazard ratio, 4.6; 95% confidence interval, 1.7-12.1 [validation]). The imaging signature remained an independent predictor of PFS in multivariable analysis adjusting for stage, human papillomavirus status, and smoking history. Conclusion: The proposed imaging signature allows more accurate prediction of disease progression and, if prospectively validated, may refine OPSCC patient selection for risk-adaptive therapy.

A Radiation Oncology-Specific Automated Trigger Indicator Tool for High-Risk, Near-Miss Safety Events

PURPOSE

Error detection in radiation oncology relies heavily on voluntary reporting, and many adverse events and near misses likely go undetected. Trigger tools use existing data in patient charts to identify otherwise-unaccounted-for events and have been successfully employed in other areas of medicine. We developed an automated radiation oncology-specific trigger tool and validated it against near-miss data from a high-volume incident learning system (ILS).

METHODS AND MATERIALS

Twenty triggers were derived from an electronic radiation oncology information system. Data from the systems over an approximately 3.5-year period were split randomly into training and test sets. The probability of a high-grade (grade 3-4) near miss for each treatment course in the training set was estimated using a regularized logistic regression model. The predictive model was applied to the test set. Records for 25 flagged treatment courses with an ILS entry were reviewed to explore the association between triggers and near misses, and 25 flagged courses without an ILS entry were reviewed to detect unreported near misses.

RESULTS

Of the 3159 treatment courses analyzed, 357 had a grade 3 to 4 ILS entry; 2210 courses composed the training set, and the test set had 949 courses. Areas under the curve on the training and test sets were 0.650 and 0.652, respectively. Of 20 triggers, 9 reached statistical significance on univariate analysis. Fifty percent of the 25 treatment courses in the test set with the highest predicted likelihood of a high-grade near miss with an ILS entry had a direct relationship between the triggers and the near miss. Review of the 25 treatment courses with the highest predicted likelihood of high-grade near miss without an ILS entry found 2 unreported near-miss events.

CONCLUSIONS

The radiation oncology-specific automated trigger tool performed modestly and identified additional treatment courses with near-miss events. Radiation oncology trigger tools deserve further exploration.

Predictors of Respiratory Decline Following Stereotactic Ablative Radiotherapy to Multiple Lung Tumors

INTRODUCTION

Stereotactic ablative radiotherapy (SABR) is highly effective at controlling early stage primary lung cancer and lung metastases. Although previous studies have suggested that treating multiple lung tumors with SABR is safe, post-treatment changes in respiratory function have not been analyzed in detail.

PATIENTS AND METHODS

We retrospectively identified patients with 2 or more primary lung cancers or lung metastases treated with SABR and analyzed clinical outcomes and predictors of toxicity. We defined a composite respiratory decline endpoint to include increased oxygen requirement, increased dyspnea scale, or death from respiratory failure not owing to disease progression.

RESULTS

A total of 86 patients treated with SABR to 203 lung tumors were analyzed. A total of 21.8% and 41.8% of patients developed composite respiratory decline at 2 and 4 years, respectively. When accounting for intrathoracic disease progression, 12.7% of patients developed composite respiratory decline at 2 years. Of the patients, 7.9% experienced grade 2 or greater radiation pneumonitis. No patient- or treatment-related factor predicted development of respiratory decline. The median overall survival was 46.9 months, and the median progression-free survival was 14.8 months. The cumulative incidence of local failure was 9.7% at 2 years.

CONCLUSION

Although our results confirm that SABR is an effective treatment modality for patients with multiple lung tumors, we observed a high rate of respiratory decline after treatment, which may be owing to a combination of treatment and disease effects. Future studies may help to determine ways to avoid pulmonary toxicity from SABR.

Automated Survival Prediction in Metastatic Cancer Patients Using High-Dimensional Electronic Medical Record Data

BACKGROUND

Oncologists use patients' life expectancy to guide decisions and may benefit from a tool that accurately predicts prognosis. Existing prognostic models generally use only a few predictor variables. We used an electronic medical record dataset to train a prognostic model for patients with metastatic cancer.

METHODS

The model was trained and tested using 12 588 patients treated for metastatic cancer in the Stanford Health Care system from 2008 to 2017. Data sources included provider note text, labs, vital signs, procedures, medication orders, and diagnosis codes. Patients were divided randomly into a training set used to fit the model coefficients and a test set used to evaluate model performance (80%/20% split). A regularized Cox model with 4126 predictor variables was used. A landmarking approach was used due to the multiple observations per patient, with t0 set to the time of metastatic cancer diagnosis. Performance was also evaluated using 399 palliative radiation courses in test set patients.

RESULTS

The C-index for overall survival was 0.786 in the test set (averaged across landmark times). For palliative radiation courses, the C-index was 0.745 (95% confidence interval [CI] = 0.715 to 0.775) compared with 0.635 (95% CI = 0.601 to 0.669) for a published model using performance status, primary tumor site, and treated site (two-sided P < .001). Our model's predictions were well-calibrated.

CONCLUSIONS

The model showed high predictive performance, which will need to be validated using external data. Because it is fully automated, the model can be used to examine providers' practice patterns and could be deployed in a decision support tool to help improve quality of care.

Increases in Serial Pretreatment 18F-FDG PET-CT Metrics Predict Survival in Early Stage Non-Small Cell Lung Cancer Treated With Stereotactic Ablative Radiation Therapy

Purpose

Quantitative changes in positron emission tomography with computed tomography imaging metrics over serial scans may be predictive biomarkers. We evaluated the relationship of pretreatment metabolic tumor growth rate (MTGR) and standardized uptake value velocity (SUVV) with disease recurrence or death in patients with early-stage non-small cell lung cancer treated with stereotactic ablative radiation therapy (SABR).

Methods and Materials

Under institutional review board approval, we retrospectively identified patients who underwent positron emission tomography with computed tomography at diagnosis and staging and simulation for SABR. Two cohorts underwent SABR between November 2005 to October 2012 (discovery) and January 2012 to April 2016 (validation). MTGR and SUVV were calculated as the daily change in metabolic tumor volume and maximum standardized uptake value, respectively. Cox proportional hazard models identified predictors of local, regional, and distant recurrence and death for the combined cohort. MTGR and SUVV thresholds dichotomizing risk of death in the discovery cohort were applied to the validation cohort.

Results

A total of 152 lesions were identified in 143 patients (92 lesions in 83 discovery cohort patients). In multivariable models, increasing MTGR trended toward increased hazard of distant recurrence (hazard ratio, 6.98; 95% confidence interval, 0.67-72.61; P = .10). In univariable models, SUVV trended toward risk of death (hazard ratio, 11.8, 95% confidence interval, 0.85-165.1, P = .07). MTGR greater than 0.04 mL/d was prognostic of decreased survival in discovery (P = .048) and validation cohorts (P < .01).

Conclusions

MTGR greater than 0.04 mL/d is prognostic of death in patients with non-small cell lung cancer treated with SABR. Increasing SUVV trends, nonsignificantly, toward increased risk of recurrence and death. MTGR and SUVV may be candidate imaging biomarkers to study in trials evaluating systemic therapy with SABR for patients at high risk of out-of-field recurrence.