Search

Background: Artificial intelligence (AI) and deep learning have shown great potential in streamlining clinical tasks. However, most studies remain confined to in silico validation in small internal cohorts, without external validation or data on real-world clinical utility. We developed a strategy for the clinical validation of deep learning models for segmenting primary non-small-cell lung cancer (NSCLC) tumours and involved lymph nodes in CT images, which is a time-intensive step in radiation treatment planning, with large variability among experts., Methods: In this observational study, CT images and segmentations were collected from eight internal and external sources from the USA, the Netherlands, Canada, and China, with patients from the Maastro and Harvard-RT1 datasets used for model discovery (segmented by a single expert). Validation consisted of interobserver and intraobserver benchmarking, primary validation, functional validation, and end-user testing on the following datasets: multi-delineation, Harvard-RT1, Harvard-RT2, RTOG-0617, NSCLC-radiogenomics, Lung-PET-CT-Dx, RIDER, and thorax phantom. Primary validation consisted of stepwise testing on increasingly external datasets using measures of overlap including volumetric dice (VD) and surface dice (SD). Functional validation explored dosimetric effect, model failure modes, test-retest stability, and accuracy. End-user testing with eight experts assessed automated segmentations in a simulated clinical setting., Findings: We included 2208 patients imaged between 2001 and 2015, with 787 patients used for model discovery and 1421 for model validation, including 28 patients for end-user testing. Models showed an improvement over the interobserver benchmark (multi-delineation dataset; VD 0·91 [IQR 0·83-0·92], p=0·0062; SD 0·86 [0·71-0·91], p=0·0005), and were within the intraobserver benchmark. For primary validation, AI performance on internal Harvard-RT1 data (segmented by the same expert who segmented the discovery data) was VD 0·83 (IQR 0·76-0·88) and SD 0·79 (0·68-0·88), within the interobserver benchmark. Performance on internal Harvard-RT2 data segmented by other experts was VD 0·70 (0·56-0·80) and SD 0·50 (0·34-0·71). Performance on RTOG-0617 clinical trial data was VD 0·71 (0·60-0·81) and SD 0·47 (0·35-0·59), with similar results on diagnostic radiology datasets NSCLC-radiogenomics and Lung-PET-CT-Dx. Despite these geometric overlap results, models yielded target volumes with equivalent radiation dose coverage to those of experts. We also found non-significant differences between de novo expert and AI-assisted segmentations. AI assistance led to a 65% reduction in segmentation time (5·4 min; p<0·0001) and a 32% reduction in interobserver variability (SD; p=0·013)., Interpretation: We present a clinical validation strategy for AI models. We found that in silico geometric segmentation metrics might not correlate with clinical utility of the models. Experts' segmentation style and preference might affect model performance., Funding: US National Institutes of Health and EU European Research Council., Competing Interests: Declaration of interests AH reports consultancy fees for Altis Labs; and owning stock in Altis Labs. HJWLA reports being a scientific adviser and shareholder for Onc.Ai, Love Health, Health-AI, and Bristol Myers Squibb. RHM reports a research grant from ViewRay; honoraria from NewRT and ViewRay; is on an advisory board for AstraZeneca; and reports travel expenses from NewRT. DEK reports honoraria from RefleXion; and is a consultant and on an advisory board for Genentech. All other authors declare no competing interests., (Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.)

Objective: Neurologic death is the most serious consequence of intracranial disease among patients with brain metastases. Identifying patients with brain metastases at increased risk of neurologic death can improve care and guide further research. We sought to delineate factors predictive of neurologic death among patients with brain metastases., Methods: We identified 1218 patients with newly diagnosed brain metastases managed at Brigham and Women's Hospital/Dana-Farber Cancer Institute from 2008-2015. Factors predictive of neurologic death were assessed via univariable and multivariable Fine and Gray competing risks regression., Results: On multivariable analysis, neurologic death was associated with number of brain metastases (hazard ratio [HR] 1.01 per 1 metastasis increase, 95% confidence interval [CI] 1.01-1.02, P < 0.001) and 3 primary tumor sites (reference=non-small cell lung cancer): melanoma (HR 4.67, 95% CI 3.27-6.68, P < 0.001), small cell lung cancer (HR 2.33, 95% CI 1.47-3.68, P < 0.001), and gastrointestinal cancer (HR 2.21, 95% CI 1.28-3.82, P = 0.005). Conversely, a reduction in neurologic death was found in patients with good Karnofsky performance status (90-100 vs. 30-80, HR 0.67, 95% CI 0.48-0.95, P = 0.03) and progressive extracranial metastases at diagnosis of intracranial disease (HR 0.50, 95% CI 0.38-0.67, P = 0.001). Among patients with breast primaries, HER2+ patients displayed increased neurologic death relative to the reference of HR+/HER2- (univariable analysis only: HR 2.41, 95% CI 1.00-5.84, P = 0.05)., Conclusions: Patients with melanoma, small cell lung cancer, gastrointestinal cancer, and HER2+ breast cancer primaries, as well as greater intracranial versus extracranial disease burden, harbor significant risk of neurologic death. Future research investigating novel intracranial approaches should focus on these populations., (Copyright © 2022 Elsevier Inc. All rights reserved.)

Introduction: The diagnosis and treatment of lung cancer is challenged by complex diagnostic pathways and fragmented care that can lead to care disparities for vulnerable patients., Methods: A multi-institutional, multidisciplinary conference was convened to address the complexity of lung cancer care particularly in patients at high-risk for treatment delay. The resulting care delivery model, called the Lung Cancer Strategist Program (LCSP), was led by a thoracic-trained advanced practice provider (APP) with emphasis on expedited surgery and early oncologic consultation in the assessment of a newly diagnosed suspicious lung nodule. We performed a retrospective review to evaluate care efficiency and oncologic outcomes in the first 100 LCSP patients compared to 100 concurrent patients managed via routine surgical referral., Results: In the 78 LCSP and 41 routine referral patients managed via nodule surveillance, LCSP patients had a shorter time from suspicious finding to work-up (3 vs. 26 days, p < 0.001) and to surveillance decision (12.5 vs. 39 days, p < 0.001). In the 22 LCSP and 59 routine referral patients treated for intrathoracic malignancy, LCSP patients had fewer hospital visits (4 vs 6, p < 0.001), clinicians seen (1.5 vs. 2, p = 0.08), and diagnostic studies (4 vs 5, p = 0.01) with a shorter time to diagnosis (30.5 vs. 48 days, p = 0.02) and treatment (40.5 vs. 68.5 days, p = 0.02)., Conclusions: Patient triage through a thoracic-trained APP in consultation with surgical, medical, and radiation oncology facilitates rapid assessment of benign versus malignant lesions with reduced time to diagnosis and treatment, even among patients at high-risk for treatment delay., (Copyright © 2021. Published by Elsevier Inc.)

Background: Prognostic estimates for patients with brain metastases (BM) stem from younger, healthier patients enrolled in clinical trials or databases from academic centers. We characterized population-level prognosis in elderly patients with BM., Methods: Using Surveillance, Epidemiology, and End Results (SEER)-Medicare data, we identified 9882 patients ≥65 years old with BM secondary to lung, breast, skin, kidney, esophageal, colorectal, and ovarian primaries between 2014 and 2016. Survival was assessed by primary site and evaluated with Cox regression., Results: In total, 2765 versus 7117 patients were diagnosed with BM at primary cancer diagnosis (synchronous BM, median survival = 2.9 mo) versus thereafter (metachronous BM, median survival = 3.4 mo), respectively. Median survival for all primary sites was ≤4 months, except ovarian cancer (7.5 mo). Patients with non-small-cell lung cancer (NSCLC) receiving epidermal growth factor receptor (EGFR)- or anaplastic lymphoma kinase (ALK)-based therapy for synchronous BM displayed notably better median survival at 12.5 and 20.1 months, respectively, versus 2.8 months exhibited by other patients with NSCLC; survival estimates in melanoma patients based on receipt of BRAF/MEK therapy versus not were 6.7 and 2.8 months, respectively. On multivariable regression, older age, greater comorbidity, and type of managing hospital were associated with poorer survival; female sex, higher median household income, and use of brain-directed stereotactic radiation, neurosurgical resection, or systemic therapy (versus brain-directed non-stereotactic radiation) were associated with improved survival (all P < 0.05)., Conclusions: Elderly patients with BM have a poorer prognosis than suggested by prior algorithms. If prognosis is driven by systemic and not intracranial disease, brain-directed therapy with potential for significant toxicity should be utilized cautiously., (© The Author(s) 2020. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)

Background: Prior literature has suggested synergy between immune checkpoint therapy (ICT) and radiotherapy (RT) for the treatment of brain metastases (BrM), but to the authors' knowledge the optimal timing of therapy to maximize this synergy is unclear., Methods: A total of 199 patients with melanoma and non-small cell lung cancer with BrM received ICT and RT between 2007 and 2016 at the study institution. To reduce selection biases, individual metastases were included only if they were treated with RT within 90 days of ICT. Concurrent treatment was defined as RT delivered on the same day as or in between doses of an ICT course; all other treatment was considered to be nonconcurrent. Multivariable Cox proportional hazards models were used to assess time to response and local disease recurrence on a per-metastasis basis, using a sandwich estimator to account for intrapatient correlation., Results: The final cohort included 110 patients with 340 BrM, with 102 BrM treated concurrently and 238 BrM treated nonconcurrently. Response rates were higher with the use of concurrent treatment (70% vs 47%; P < .001), with correspondingly lower rates of progressive disease (5% vs 26%; P < .001). On multivariable analysis, concurrent treatment was found to be associated with improved time to response (hazard ratio, 1.76; 95% CI, 1.18-2.63 [P = .006]) and decreased local recurrence (hazard ratio, 0.42; 95% CI, 0.23-0.78 [P = .006]). This effect appeared to be greater for melanoma than for non-small cell lung cancer, although interaction tests were not statistically significant. Only 1 of 103 metastases which had a complete response later developed disease progression., Conclusions: Concurrent RT and ICT may improve response rates and decrease local recurrence of brain metastases compared with treatment that was nonconcurrent but delivered within 90 days. Further study of this combination in prospective, randomized trials is warranted., (© 2020 American Cancer Society.)

Background: Pemetrexed is a folate antimetabolite used in the management of advanced adenocarcinoma of the lung. We sought to assess the impact of pemetrexed on intracranial disease control and radiation-related toxicity among patients with adenocarcinoma of the lung who received stereotactic radiation for brain metastases., Materials/methods: We identified 149 patients with adenocarcinoma of the lung and newly diagnosed brain metastases without a targetable mutation receiving stereotactic radiation. Kaplan-Meier plots and Cox regression were employed to assess whether use of pemetrexed was associated with intracranial disease control and radiation necrosis., Results: Among the entire cohort, 105 patients received pemetrexed while 44 did not. Among patients who were chemotherapy-naïve, use of pemetrexed (n = 43) versus alternative regimens after stereotactic radiation (n = 24) was associated with a reduced likelihood of developing new brain metastases (HR 0.42, 95% CI 0.22-0.79, p = 0.006) and a reduced need for salvage brain-directed radiation therapy (HR 0.36, 95% CI 0.18-0.73, p = 0.005). Pemetrexed use was associated with increased radiographic necrosis. (HR 2.70, 95% CI 1.09-6.70, p = 0.03)., Conclusions: Patients receiving pemetrexed after brain-directed stereotactic radiation appear to benefit from improved intracranial disease control at the possible expense of radiation-related radiographic necrosis. Whether symptomatic radiation injury occurs more frequently in patients receiving pemetrexed requires further study., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Background and Purpose: There is concern that patients with collagen vascular disease (CVD) are at higher risk of developing radiation toxicity. We analyzed radiation toxicities in patients with intrathoracic malignancy and CVD treated using modern radiotherapy., Materials and Methods: This single-institution retrospective study included 31 patients with CVD and 825 patients without CVD treated from 1998 to 2014. Radiation esophagitis (RE) and radiation pneumonitis (RP) were scored by RTOG scales. RE was analyzed with logistic regression and RP with Cox regression., Results: CVD patients experienced similar grade ≥3 RE compared to control patients (23% vs. 19%, p = 0.64) but more grade ≥3 RP (26% vs. 10%, p = 0.01). There was no significant association between CVD subtype and toxicities. In multivariate analysis, CVD and lung V20 >30% were associated with grade ≥3 RP. We identified V20 ≤30%, V5 ≤50%, and MLD ≤18 Gy as dose thresholds in patients with CVD. CVD patients with mild severity disease and only 1 organ system involved were at low risk for RP., Conclusions: Patients with CVD may be at higher risk of RP. However, CVD patients may be offered curative thoracic RT with particular attention to risk-reduction strategies and maintaining recommended dose constraints as described in this study., (Copyright © 2017 Elsevier B.V. All rights reserved.)

Background: We analyzed outcomes after lung stereotactic body radiotherapy (SBRT) for early-stage non-small cell lung-carcinoma (NSCLC) by histology and KRAS genotype., Patients and Methods: We included 75 patients with 79 peripheral tumors treated with SBRT (18 Gy × 3 or 10 to 12 Gy × 5) at our institution from 2009 to 2012. Genotyping for KRAS mutations was performed in 10 patients. Outcomes were analyzed by the Kaplan-Meier method/Cox regression, or cumulative incidence method/Fine-Gray analysis., Results: The median patient age was 74 (range, 46 to 93) years, and Eastern Cooperative Oncology Group performance status was 0 to 1 in 63%. Tumor histology included adenocarcinoma (44%), squamous cell carcinoma (25%), and NSCLC (18%). Most tumors were T1a (54%). Seven patients had KRAS-mutant tumors (9%). With a median follow-up of 18.8 months among survivors, the 1-year estimate of overall survival was 88%, cancer-specific survival (CSS) 92%, primary tumor control 94%, and freedom from recurrence (FFR) 67%. In patients with KRAS-mutant tumors, there was a significantly lower tumor control (67% vs. 96%; P = .04), FFR (48% vs. 69%; P = .03), and CSS (75% vs. 93%; P = .05). On multivariable analysis, histology was not associated with outcomes, but KRAS mutation (hazard ratio, 10.3; 95% confidence interval, 2.3-45.6; P = .0022) was associated with decreased CSS after adjusting for age., Conclusion: In this SBRT series, histology was not associated with outcomes, but KRAS mutation was associated with lower FFR on univariable analysis and decreased CSS on multivariable analysis. Because of the small sample size, these hypothesis-generating results need to be studied in larger data sets., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Context: Distressing symptoms interfere with the quality of life in patients with lung cancer. Algorithm-based clinical decision support (CDS) to improve evidence-based management of isolated symptoms seems promising, but no reports yet address multiple symptoms., Objectives: This study examined the feasibility of CDS for a Symptom Assessment and Management Intervention targeting common symptoms in patients with lung cancer (SAMI-L) in ambulatory oncology. The study objectives were to evaluate completion and delivery rates of the SAMI-L report and clinician adherence to the algorithm-based recommendations., Methods: Patients completed a web-based symptom assessment and SAMI-L created tailored recommendations for symptom management. Completion of assessments and delivery of reports were recorded. Medical record review assessed clinician adherence to recommendations. Feasibility was defined as 75% or higher report completion and delivery rates and 80% or higher clinician adherence to recommendations. Descriptive statistics and generalized estimating equations were used for data analyses., Results: Symptom assessment completion was 84% (95% CI=81-87%). Delivery of completed reports was 90% (95% CI=86-93%). Depression (36%), pain (30%), and fatigue (18%) occurred most frequently, followed by anxiety (11%) and dyspnea (6%). On average, overall recommendation adherence was 57% (95% CI=52-62%) and was not dependent on the number of recommendations (P=0.45). Adherence was higher for anxiety (66%; 95% CI=55-77%), depression (64%; 95% CI=56-71%), pain (62%; 95% CI=52-72%), and dyspnea (51%; 95% CI=38-64%) than for fatigue (38%; 95% CI=28-47%)., Conclusion: The CDS systems, such as SAMI-L, have the potential to fill a gap in promoting evidence-based care., (Copyright © 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.)

Purpose: Only 2% to 5% of adult patients with cancer enroll onto clinical trials. We assessed simultaneously characteristics of patients and their physicians that may be independently associated with participation., Methods: CanCORS, a National Cancer Institute (NCI) -funded population-based observational cohort study of newly diagnosed patients with lung and colorectal cancers, sampled patients across five geographic areas, five health care delivery systems, and 15 Veterans Administration hospitals. We linked patient survey and medical record data with physician survey data to examine correlates of trial enrollment., Results: Among 9,901 patients, 5.3% enrolled onto trials. Of the 9,901 patients, we linked 6,506 patients to one medical oncologist, surgeon, or radiation oncologist (physicians, N = 1,325) who responded to the physician survey and was considered their primary cancer clinician decision maker. Patient age, race, disease stage, geographic region, and health insurance were independently associated with trial enrollment. Physician factors independently associated with patient trial enrollment were being a medical oncologist, practicing at an NCI-designated cancer center, taking the lead in discussing trials with patients, and receiving increased income from trial enrollment. After simultaneously adjusting for patient and physician characteristics, only being a physician practicing at an NCI-designated cancer center (odds ratio [OR], 1.65; 95% CI, 1.19 to 2.27) and patient female sex (OR, 1.36; 95% CI, 1.10 to 1.68), age > 70 versus < 50 years (OR, 0.28; 95% CI, 0.16 to 0.48), and advanced disease (OR, 1.85; 95% CI, 1.45 to 2.37) remained independently associated with trial enrollment., Conclusion: Both practice environment and patient clinical and demographic characteristics are associated with cancer clinical trial enrollment; simultaneous intervention may be required when trying to increase enrollment rates.

Background: The research goals of the Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium are to determine how characteristics and beliefs of patients, providers, and health care organizations influence the treatments and outcomes of individuals with newly diagnosed lung and colorectal cancers. As CanCORS results will inform national policy, it is important to know how they generalize to the United States population with these cancers., Research Design: This study assessed the representativeness of the CanCORS cohort of 10,547 patients with lung cancer (LC) or colorectal cancer (CRC) enrolled between 2003 and 2005. We compared characteristics (sex, race, age, and disease stage) with the Surveillance, Epidemiology, and End Results (SEER) population of 234,464 patients with new onset of these cancers during the CanCORS recruitment period., Results: The CanCORS sample is well matched to the SEER Program for both cancers. In CanCORS, 41% LC/47% CRC were female versus 47% LC/49% CRC in SEER. African American, Hispanic, and Asian cases differed by no more than 5 percentage points between CanCORS and SEER. The SEER population is slightly older, with the percentage of patients older than 75 years 33.1% LC/37.3% CRC in SEER versus 26.9% LC/29.4% in CanCORS, and also has a slightly higher proportion of early stage patients. We also found that the CanCORS cohort was representative within specific SEER regions that map closely to CanCORS sites., Conclusions: This study demonstrates that the CanCORS Consortium was successful in enrolling a demographically representative sample within the CanCORS regions.

Background: Chemotherapy for metastatic lung or colorectal cancer can prolong life by weeks or months and may provide palliation, but it is not curative., Methods: We studied 1193 patients participating in the Cancer Care Outcomes Research and Surveillance (CanCORS) study (a national, prospective, observational cohort study) who were alive 4 months after diagnosis and received chemotherapy for newly diagnosed metastatic (stage IV) lung or colorectal cancer. We sought to characterize the prevalence of the expectation that chemotherapy might be curative and to identify the clinical, sociodemographic, and health-system factors associated with this expectation. Data were obtained from a patient survey by professional interviewers in addition to a comprehensive review of medical records., Results: Overall, 69% of patients with lung cancer and 81% of those with colorectal cancer did not report understanding that chemotherapy was not at all likely to cure their cancer. In multivariable logistic regression, the risk of reporting inaccurate beliefs about chemotherapy was higher among patients with colorectal cancer, as compared with those with lung cancer (odds ratio, 1.75; 95% confidence interval [CI], 1.29 to 2.37); among nonwhite and Hispanic patients, as compared with non-Hispanic white patients (odds ratio for Hispanic patients, 2.82; 95% CI, 1.51 to 5.27; odds ratio for black patients, 2.93; 95% CI, 1.80 to 4.78); and among patients who rated their communication with their physician very favorably, as compared with less favorably (odds ratio for highest third vs. lowest third, 1.90; 95% CI, 1.33 to 2.72). Educational level, functional status, and the patient's role in decision making were not associated with such inaccurate beliefs about chemotherapy., Conclusions: Many patients receiving chemotherapy for incurable cancers may not understand that chemotherapy is unlikely to be curative, which could compromise their ability to make informed treatment decisions that are consonant with their preferences. Physicians may be able to improve patients' understanding, but this may come at the cost of patients' satisfaction with them. (Funded by the National Cancer Institute and others.).

Goals of Work: The National Cancer Institute's Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium is conducting a population-based study of newly diagnosed patients with lung and colorectal cancer to describe the experience of persons living with cancer and to understand which barriers present the most significant obstacles to their receipt of appropriate care. The keystone to this effort is the baseline patient survey administered approximately 4 months after diagnosis., Patients and Methods: We developed a survey to obtain information from patients newly diagnosed with lung and colorectal cancer about their personal characteristics, decision making, experience of care, and outcomes. We conducted a pilot study to evaluate the feasibility of a lengthy and clinically detailed interview in a convenience sample of patients within 8 months of diagnosis (n=71)., Main Results: The median length of the interviews was 75 min for patients with lung cancer (range 43-130) and 82 min for patients with colorectal cancer (range 46-119). Most patients had received some form of treatment for their cancer: 66.1% had undergone surgery, 28.2% had received radiation therapy, and 54.9% were treated with chemotherapy. In addition, 26.7% reported their overall health was less than 70 on a 0-100 scale, demonstrating that patients with substantial health impairment were able to complete the survey., Conclusions: A clinically detailed survey of newly diagnosed lung and colorectal cancer patients is feasible. A modified version of this survey is being fielded by the CanCORS Consortium and should provide much needed population-based data regarding patients' experiences across the continuum of cancer care and their outcomes.