Non-small Cell Lung Research Articles

Moderating effects of organizational minority representation on racial and ethnic differences in early palliative care receipt among patients diagnosed with advanced cancers in 2010-2019 in the US.

52 Background: Early palliative care (PC) is recommended for advanced cancers, but differential use has been documented by race and ethnicity. We examined whether racial and ethnic differences in early PC use were moderated by organization-level proportions of minority patients served. Methods: We identified patients aged ≥65.5 years diagnosed with advanced-stage breast, colorectal, non-small cell lung (NSCL), small cell lung (SCL), pancreas, and prostate cancers in 2010-2019 with ≥6 months survival and continuous fee-for-service coverage from the SEER-Medicare data. Early PC was identified by PC encounter diagnosis codes or hospice and palliative medicine (HPM) specialty codes within 90 days post-diagnosis or up to first hospice admission date. Treating physicians and organizations were assigned based on the plurality of visits within 180 days surrounding diagnosis. Using patient race and ethnicity information, organizations were classified based on the percent of racial and ethnic minorities patients (Low<10%, Medium 10-19%, and High Minority-Serving ≥20%). Multivariable logistic models assessed interacting effects of organizations’ minority representation and individual race and ethnicity on early PC receipt, adjusting for patient characteristics. Results: Among 102,060 patients treated at 8,039 organizations, the percent receiving early PC increased from 1.44% in 2010 to 10.30% in 2019. High Minority-Serving organizations were larger size and more likely to employ a HPM specialist and to treat a higher proportion of patients in metropolitan but lower socioeconomic areas, and were dually Medicare-Medicaid eligible. After adjusting for patient characteristics, Hispanic patients were 1.6 percentage points (ppts, 95%CI=-3.1, -0.1) less likely to receive early PC than Non-Hispanic (NH) White patients in Low Minority-Serving organizations. In High Minority-Serving organizations, Hispanic and NH Black patients were 0.8 ppts (95%CI=0.1, 1.6) and 1.6 ppts (95%CI=1.0, 2.3) more likely to receive early PC than NH White patients. (Table). Conclusions: Racial and ethnic minorities were more likely to receive early PC than NH White at High Minority-Serving organizations but not Medium/Low Minority-Serving organizations. System-level factors may be modified to narrow racial and ethnic differences in early PC use. Adjusted difference in early palliative care by race and ethnicity. Organization Minority Representation Individual Adjusted Probability Difference (95% CI) Low NH White Ref Hispanic -1.6 (-3.1, -0.1) NH Black 0.01 (-1.7, 1.7) Other/Unknown 1.1 (-1.2, 3.4) Medium NH White Ref Hispanic -0.6 (-1.7, 0.5) NH Black -0.8 (-1.7, 0.2) Other/Unknown 0.4 (-0.9, 1.7) High Non-Hispanic White Ref Hispanic 0.8 (0.1, 1.6) NH Black 1.6 (1.0, 2.3) Other/Unknown -0.3 (-1.1, 0.4)

Uptake of and factors associated with telemedicine use for palliative care among patients diagnosed with advanced cancers during the onset of COVID-19.

53 Background: The COVID-19 pandemic propelled a rapid shift to telemedicine use, including patients with advanced cancers needing palliative care (PC). Yet, little is known about disparities in telemedicine use for PC. We leveraged population-based data to examine the trend and factors associated with telemedicine use for PC among Medicare beneficiaries diagnosed with advanced cancers during the onset of COVID-19 pandemic. Methods: We identified patients aged ≥65.5 years diagnosed with advanced-stage breast, colorectal, non-small cell lung (NSCL), small cell lung (SCL), pancreas, and prostate cancers in 2015-2019 from the SEER-Medicare data. We included those with continuous fee-for-service coverage and had any PC claims from diagnosis to death, hospice admission, or by 12/31/2020 (whichever came first). Telemedicine was identified using procedure codes, modifiers, and place of service codes on PC claims. We described the percent of PC recipients using telemedicine each quarter in 2019-2020. Multivariable logistic models assessed patient and provider characteristics associated with telemedicine use among PC recipients in 2020. Results: Telemedicine use for PC was 0.24% to 0.48% across 4 quarters in 2019, which increased to 16.3% in Q1 and stabilized at about 30% in Q2-Q4 in 2020. Among 2,945 PC recipients in 2020, 22.5% used telemedicine. Older patients (75-84 years and 85+ years) were 4.4 percentage points (ppts) and 8.6 ppts (p-values<0.05) less likely than patients ages 65-74 years to use telemedicine. Patients who were female (4.4 ppts), residing in higher socioeconomic status areas (Yost Quintile 5 vs 1: 7.7 ppts), treated by female providers (4.1 ppts), in hospital system (3.7 ppts vs. individual physician office) and organizations with hospice and palliative care specialists (8.9 ppts) were more likely to use telemedicine for PC (all p-values<0.05). (Table). Conclusions: Telemedicine use for PC surged during the onset of the COVID-19 pandemic, with considerable disparities by patient sociodemographic and provider characteristics. These disparities suggest potential barriers related to patients’ technology literacy and internet access, provider awareness, and healthcare organization resources. Future mixed-method studies need to elucidate these barriers to design interventions at patient, provider, and system levels to ensure effective PC delivery. Selected factors associated with telemedicine use for palliative care. Absolute Probability Difference (95%CI) Age group (ref: 65-74) 75-84 -4.4 (-7.6, -1.2) 85+ -8.6 (-13.4, -3.7) Female (ref: Male) 4.4 (0.9, 7.9) Yost quintile (ref : Q1-Q2) Q3-Q4 2.2 (-1.7, 6.1) Q5 7.7 (3.3, 12.1) Hospital system (ref: Individual physician office) 3.7 (0.6, 6.7) Provider: Female (ref: Male) 4.1 (0.7, 7.5) Organization with hospice and palliative care specialists 8.9 (5.4, 12.4)

Is federal housing assistance associated with earlier-stage cancer diagnosis? Evidence from the new SEER-Medicare and US Department of Housing and Urban Development data linkage.

85 Background: Despite growing awareness that stable and affordable housing is a key social determinant of health, little research has examined the role of federal housing assistance in cancer outcomes. This study uses a novel data linkage to examine the association of federal housing assistance (i.e. various rent subsidy programs) on stage at diagnosis for the 4 most common cancers in the US. Methods: Individuals aged 66-95 years newly diagnosed with breast, colorectal, non-small cell lung (NSCLC) or prostate cancer in 2006-2019 with fee-for-service coverage were selected from SEER-Medicare Housing and Urban Development (HUD) administrative data. After identifying individuals with housing assistance at diagnosis, controls without housing assistance at diagnosis were propensity score matched in a 3:1 ratio by cancer site, age, sex, race/ethnicity, marital status, registry, year of diagnosis, area-level Yost socioeconomic status index, dual Medicaid eligibility, rural/urban residence, reason for Medicare entitlement, and pre-diagnosis comorbidity. Associations of housing assistance and diagnosis stage (SEER summary stage: localized vs. regional/distant) were examined with separate logistic regression models by cancer type. Results: A total of 7335 individuals with breast, 5660 with colorectal, 7399 with NSCL and 4309 with prostate cancer received housing assistance at diagnosis (Table). Compared to matched controls, a smaller percentage of individuals with housing assistance were diagnosed with regional/distant breast (36% vs. 33%), colorectal (59% vs. 57%), and NSCL (71% vs. 69%) cancers. Individuals receiving housing assistance were less likely to be diagnosed with later-stage breast (OR: 0.87; 95% CI: 0.83, 0.92) and NSCL (OR: 0.93; 95% CI: 0.88, 0.99) cancers. Housing assistance was not significantly associated with stage for individuals with colorectal or prostate cancers. Conclusions: Federal housing assistance is associated with earlier-stage breast and NSCL cancer diagnosis, highlighting its potential role in mitigating the adverse effects of housing insecurity on cancer outcomes. Housing assistance and cancer diagnosis stage. No Housing Assistance at Dx,N (%) Housing Assistance at Dx,N (%) OR [95% CI], P Breast Localized 14291 (64) 4890 (67) 1.0 (ref) Regional/Distant 8179 (36) 2445 (33) 0.87 [0.83, 0.92] P<0.001 Colorectal Localized 7046 (42) 2422 (43) 1.0 (ref) Regional/Distant 9934 (59) 3238 (57) 0.95 [0.89, 1.01] P=0.09 NSCLC Localized 6572 (30) 2298 (31) 1.0 (ref) Regional/Distant 15625 (71) 5101 (69) 0.93 [0.88, 0.99] P=0.02 Prostate Localized 10293 (80) 3401 (79) 1.0 (ref) Regional/Distant 2634 (20) 908 (21) 1.04 [0.96, 1.14] P=0.33 Dx=Diagnosis; OR=Odds Ratio; CI=Confidence Interval.

Abstract C113: Parabens promote cell growth in cancer cell lines associated with health disparities

Abstract Parabens are endocrine-disrupting chemicals used as preservatives to extend the shelf life of consumables, pharmaceuticals, and personal care products. Daily use of products containing parabens results in continuous exposure that can negatively impact health. It is known that parabens cause breast cancer cells to grow and express genes linked to cancer and hormone action. Our lab previously showed that parabens increase viability of and regulate estrogen-responsive gene expression in diverse luminal A breast cancer cell lines. In addition, greater effects were seen in a luminal A cell line of West African ancestry compared to one of European ancestry, suggesting that exposure to parabens in personal care products may contribute to breast cancer disparities. Therefore, the purpose of this study was to investigate whether parabens increase cell viability across a variety of cancer cell types, particularly those associated with cancer disparities. To do so, we utilized the National Cancer Institute-60 (NCI-60) Human Tumor Cell Lines screen that includes leukemia, non-small cell lung (NSCL), colon, central nervous system (CNS), melanoma, ovarian, renal, prostate, and breast cancer cell lines, plus a custom panel of pancreatic cancer cell lines. Of these, ovarian, NSCL, prostate, colon, renal, CNS, and pancreatic cancers have known disparities. We treated these cancer cell lines with biologically relevant doses of butylparaben (BP), methylparaben (MP), or propylparaben (PP) (doses ranging from 0.0012 µM – 20 µM) for 72 hours and measured cell viability. Treatment with dimethylsulfoxide or 10 nM estradiol served as negative and positive controls, respectively. In general, we observed a paraben-specific and cell-line specific effect on cell viability. MP exhibited the greatest effect with increased cell viability in 42% of the cell lines, followed by PP (38%), and BP (32%). MP and PP are among the most common parabens found in personal care products and differences in urinary levels based on race/ethnicity (e.g. higher levels in non-Hispanic Blacks/Mexican Americans compared to non-Hispanic Whites) have been reported. Breast, ovarian, and NSCL had the most cell lines (67-83%) that exhibited increased cell viability with any paraben treatment. Increased cell viability was also observed in several leukemia, colon, pancreatic, CNS, and melanoma cell lines (33-50%). Only one renal, and no prostate, cell line exhibited increased cell viability. These data suggest that paraben-mediated protumorigenic effects may promote cancer disparities beyond breast cancer. Studies are currently underway to determine the mechanisms underlying this link. One caveat of this study is that most of the cell lines in this readily available screen are of European ancestry, highlighting a need to develop more genetically diverse in vitro models for understanding the mechanistic link between paraben exposures and cancer risk/outcomes in the context of disparities. Approaching this challenge from an endocrinology and genetic ancestry perspective may help narrow the disparity gap. Citation Format: Jazma L. Tapia, Jillian C. McDonough, Loretta Erhunmwunsee, Lindsey S. Trevino. Parabens promote cell growth in cancer cell lines associated with health disparities [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr C113.

Exploring the Anticancer Potential of Phenolic nor-Triterpenes from Celastraceae Species.

To explore new compounds with antitumour activity, fifteen phenolic nor-tripterpenes isolated from Celastraceae species, Maytenus jelskii, Maytenus cuzcoina, and Celastrus vulcanicola, have been studied. Their chemical structures were elucidated through spectroscopic and spectrometric techniques, resulting in the identification of three novel chemical compounds. Evaluation on human tumour cell lines (A549 and SW1573, non-small cell lung; HBL-100 and T-47D, breast; HeLa, cervix, and WiDr, colon) revealed that three compounds, named 6-oxo-pristimerol, demethyl-zeylasteral, and zeylasteral, exhibited significant activity (GI50 ranging from 0.45 to 8.6 µM) on at least five of the cell lines tested. Continuous live cell imaging identified apoptosis as the mode of action of selective cell killing in HeLa cells. Furthermore, their effect on a drug-sensitive Saccharomyces cerevisiae strain has been investigated to deepen on their mechanism of action. In dose-response growth curves, zeylasteral and 7α-hydroxy-blepharodol were markedly active. Additionally, halo assays were conducted to assess the involvement of oxidative stress and/or mitochondrial function in the anticancer profile, ruling out these modes of action for the active compounds. Finally, we also delve into the structure-activity relationship, providing insights into how the molecular structure of these compounds influences their biological activity. This comprehensive analysis enhances our understanding of the therapeutic potential of this triterpene type and underscores its relevance for further research in this field.

Multitask Learning on Graph Convolutional Residual Neural Networks for Screening of Multitarget Anticancer Compounds.

Recently, various modern experimental screening pipelines and assays have been developed to find promising anticancer drug candidates. However, it is time-consuming and almost infeasible to screen an immense number of compounds for anticancer activity via experimental approaches. To partially address this issue, several computational advances have been proposed. In this study, we present iACP-GCR, a model based on multitask learning on graph convolutional residual neural networks with two types of shortcut connections, to identify multitarget anticancer compounds. In our architecture, the graph convolutional residual neural networks are shared by all the prediction tasks before being separately customized. The NCI-60 data set, one of the most reliable and well-known sources of experimentally verified compounds, was used to develop our model. From that data set, we collected and refined data about compounds screened across nine cancer types (panels), including breast, central nervous system, colon, leukemia, nonsmall cell lung, melanoma, ovarian, prostate, and renal, for model training and evaluation. The model performance evaluated on an independent test set shows that iACP-GCR surpasses the three advanced computational methods for multitask learning. The integration of two shortcut connection types in the shared networks also improves the prediction efficiency. We also deployed the model as a public web server to assist the research community in screening potential anticancer compounds.

Primary Care Use and 90-Day Mortality Among Older Adults Undergoing Cancer Surgery

Multimorbidity and postoperative clinical decompensation are common among older surgical patients with cancer, highlighting the importance of primary care to optimize survival. Little is known about the association between primary care use and survivorship among older adults (aged ≥65 years) undergoing cancer surgery. To examine primary care use among older surgical patients with cancer and its association with mortality. In this retrospective cohort study, data were abstracted from the electronic health record of a single health care system for older adults undergoing cancer surgery between January 1, 2017, and December 31, 2019. There were 3 tiers of stratification: (1) patients who had a primary care practitioner (PCP) (physician, nurse practitioner, or physician assistant) vs no PCP, (2) those who had a PCP and underwent surgery in the same health system (unfragmented care) vs not (fragmented care), and (3) those who had a primary care visit within 90 postoperative days vs not. Data were analyzed between August 2023 and January 2024. Primary care use after surgery for colorectal, head and neck, prostate, ovarian, pancreatic, breast, liver, renal cell, non-small cell lung, endometrial, gastric, or esophageal cancer. Postoperative 90-day mortality was analyzed using inverse propensity weighted Kaplan-Meier curves, with log-rank tests adjusted for propensity scores. The study included 2566 older adults (mean [SEM] age, 72.9 [0.1] years; 1321 men [51.5%]). Although 2404 patients (93.7%) had health insurance coverage, 743 (28.9%) had no PCP at the time of surgery. Compared with the PCP group, the no-PCP group had a higher 90-day postoperative mortality rate (2.0% vs 3.6%, respectively; adjusted P = .03). For the 823 patients with unfragmented care, 400 (48.6%) had a primary care visit within 90 postoperative days (median time to visit, 34 days; IQR, 20-57 days). Patients who had a postoperative primary care visit were more likely to be older, have a higher comorbidity burden, have an emergency department visit, and be readmitted. However, they had a significantly lower 90-day postoperative mortality rate than those who did not have a primary care visit (0.3% vs 3.3%, respectively; adjusted P = .001). These findings suggest that follow-up with primary care within 90 days after cancer surgery is associated with improved survivorship among older adults.

TIGIT expression in renal cell carcinoma infiltrating T cells is variable and inversely correlated with PD-1 and LAG3

PurposeImmune checkpoint inhibitors have revolutionized the treatment of renal cell carcinoma (RCC), but many patients do not respond to therapy and the majority develop resistant disease over time. Thus, there is increasing need for alternative immunomodulating agents. The co-inhibitory molecule T-cell immunoglobulin and ITIM domain (TIGIT) may play a role in resistance to approved immune checkpoint inhibitors and is being investigated as a potential therapeutic target. The purpose of this study was to quantify TIGIT positivity in tumor-infiltrating T cells in RCC.MethodsWe employed tissue microarrays containing specimens from primary RCC tumors, adjacent normal renal tissue, and RCC metastases to quantify TIGIT within tumor-infiltrating CD3+ T cells using quantitative immunofluorescent analysis. We also compared these results to TIGIT+ CD3+ levels in four other tumor types (melanoma, non-small cell lung, cervical, and head and neck cancers).ResultsWe did not observe significant differences in TIGIT positivity between primary RCC tumors and patient-matched metastatic samples. We found that the degree of TIGIT positivity in RCC is comparable to that in lung cancer but lower than that in melanoma, cervical, and head and neck cancers. Correlation analysis comparing TIGIT positivity to previously published, patient-matched spatial proteomic data by our group revealed a negative association between TIGIT and the checkpoint proteins PD-1 and LAG3.ConclusionOur findings support careful evaluation of TIGIT expression on T cells in primary or metastatic RCC specimens for patients who may be treated with TIGIT-targeting antibodies, as increased TIGIT positivity might be associated with a greater likelihood of response to therapy.

Bevacizumab-Based Therapies in Malignant Tumors-Real-World Data on Effectiveness, Safety, and Cost.

For this purpose, we conducted an open, observational, retrospective study involving all patients treated for solid malignant tumors in the Bucharest Institute of Oncology with "Prof. Dr. Al. Trestioreanu" with Bevacizumab-based systemic therapy, between 2017 and 2021. The study consisted of 657 treatment episodes in 625 patients (F/B = 1.62/1, with a median age of 57.6 years) which were treated for malignant tumors (majority colorectal, non-small cell lung, ovarian, and breast cancer). First-line treatment was administered in 229 patients, and the rest received Bevacizumab as second or subsequent lines of treatment. The overall response rate to Bevacizumab-based therapies was around 60-65% across all indication except for subsequent treatment lines in colorectal and ovarian cancers, where lower values were recorded (27.1%, and 31.5% respectively). Median PFS for the entire cohort was 8.2 months (95% CI 6.8-9.6), and the median OS was 13.2 months (95% CI 11.5-14.9). Usual bevacizumab-related toxicities were observed, including bleeding, hypertension, wound-healing complications, gastrointestinal perforation, other types of fistulas, septic complications, and thromboembolic events. Although the clinical benefits are undeniable, the addition of Bevacizumab to standard chemotherapy increased the overall treatment cost by 213%. Bevacizumab remains a high-cost therapy, but it can add to clinical benefits (like overall survival, progression-free survival, and response rate) when used in conjunction with standard chemotherapy. Similar results as those presented in various controlled trials are observable even on unselected cohorts of patients in the uncontrolled conditions of "real-world" oncological practice. Off-label usage is encountered in clinical practice, and this aspect should be monitored given the potential adverse effects of the therapy.

PT13 Claims-Based Machine Learning Algorithms to Predict ECOG Performance Status (ECOG) for Pan-Cancer (Prostate, Breast, Colorectal, Gastric, Non-small Cell Lung (NSCLC), and Pancreatic Cancer) Patients

PT13 Claims-Based Machine Learning Algorithms to Predict ECOG Performance Status (ECOG) for Pan-Cancer (Prostate, Breast, Colorectal, Gastric, Non-small Cell Lung (NSCLC), and Pancreatic Cancer) Patients

Time to treatment initiation for breast, prostate, non-small cell lung, and colon cancers from 2018 to 2022.

e23152 Background: The COVID-19 pandemic heavily impacted healthcare services, including cancer diagnostics and surgical management. This study aims to analyze the effects of COVID-19 on mean Time to Treatment Initiation (TTI) for breast, colon, non-small cell lung (NSCLC), and prostate cancers. Methods: Retrospective cohort study analyzing changes in mean TTI for breast, prostate, NSCLC, and colon cancer at an academic medical center from January 2018 to December 2022. “Pre-COVID” was defined as Jan 2018 – Mar 2020; “COVID” as Mar 2020 - Feb 2021; and “Post-COVID” from Mar 2021 - Dec 2022. ANOVA, chi-square, or Fisher’s exact tests were used to test for differences over the periods. Results: The cohort comprised 2014 breast, 501 colon, 1443 NSCLC, and 1355 prostate cancers. The characteristics of the study population are described in Table 1. The “clinical Stage at diagnosis 0 to 3C” was over 98% for breast among all the analyzed periods; for the colon, it decreased from 34% to 14%, while for the lung, it increased from 66% to 73%, and prostate increased from 88% to 97%. TTI was 35, 32, and 37 days (p < 0.01) for breast in the pre-COVID, COVID, and post-COVID periods, respectively. For colon, TTI was 26, 24, and 35 days in the three analyzed periods (p = 0.04), while NSCLC was 42, 46, and 50 days (p < 0.01), and prostate was 67, 57, and 64 days (p = 0.09). Conclusions: TTI was affected by COVID-19, and its effects varied according to the type of cancer. While for breast and prostate, TTI seems to be back to pre-covid period. For colon and NSCLC, the TTI is higher in the post-COVID phase. Further research to quantify the impact of delay is warranted. [Table: see text]

Evaluating generalizability of landmark randomized controlled trials in common metastatic cancers using machine learning-based emulated trials.

e13618 Background: Randomized controlled trial (RCT) results often lack generalizability to real-world oncology patients (pts). We present TrialTranslator, a machine learning (ML) framework to evaluate the generalizability of Phase III oncology RCTs by emulating trials across prognostic risk groups. Methods: TrialTranslator was applied to 11 landmark RCTs encompassing the most prevalent advanced cancers – non-small cell lung (aNSCLC), breast (mBC), prostate (mPC), and colorectal (mCRC). Flatiron Health’s EHR-derived de-identified database was used for model development and trial emulation. In Stage I, ML models (gradient boosted [GBM], random forest, support vector machine and penalized Cox) were trained to predict survival from time of metastatic diagnosis. Models used 130 longitudinal demographic, cancer-related, lab, and biomarker features. In Stage II, trial emulation, pts meeting RCT cancer and biomarker eligibility criteria were stratified into low, medium, and high-risk groups based on the top-performing ML model. We used inverse probability of treatment weighted survival analysis for each risk group to compare the median overall survival (mOS) or median progression free survival (mPFS) between experimental and control arms. Results: GBM models achieved the highest time-dependent area under the receiver operating curve (AUC) in the test set in all cancer types. For example, the 1-year AUC for the aNSCLC GBM vs the Lung Cancer Prognostic Index (validated Cox model) was 0.783 vs 0.689. ML-predicted low-risk pts typically had similar mOS or mPFS as RCT populations (Table). In emulated trials for low-risk pts, real-world survival benefit – the absolute difference in mOS or mPFS between experimental and control arms – was similar to RCT results. In contrast, for high-risk pts, survival benefits were not statistically or clinically significant. For example, in the FLAURA trial, RCT mPFS for experimental arm (osimertinib) vs control (gefitinib or erlotinib) was 18.9 mos vs 10.2 months; mPFS was 17.1 vs 10.3 months for low-risk pts and 5.1 vs 4.2 months for high-risk pts. Conclusions: ML frameworks like TrialTranslator may facilitate individual patient-level estimation of real-world treatment benefits of novel oncology agents. TrialTranslator is accessible at trialtranslator.com. [Table: see text]

Early palliative care among patients diagnosed with advanced cancers in the US (2010-2019): Trends and contribution of provider variation.

11017 Background: Early integration of palliative care (PC) is recommended for advanced cancers, but evidence of its use and the role of provider practice patterns and organizational characteristics in uptake is limited. This study examined recent trends of early PC among Medicare beneficiaries newly diagnosed with advanced cancers, and provider- and organization-variation in the receipt of early PC. Methods: We identified patients aged ≥65.5 years newly diagnosed with advanced stage breast, colorectal, non-small cell lung (NSCL), small cell lung (SCL), pancreas, and prostate cancers in 2010-2019 with ≥6 months survival and continuous fee-for-service coverage from the linked SEER-Medicare data. Early PC was identified by claims with corresponding diagnosis codes or hospice and palliative medicine provider specialty codes within 90 days post-diagnosis or up to first hospice admission date (whichever came earlier). Treating physicians and corresponding organizations (i.e., Tax Identification Number) were assigned based on the plurality of visits within 180 days after diagnosis. We described the percent of patients receiving early PC each year. Generalized linear models with physician- and organization-fixed effects evaluated variation in early PC receipt between and within physicians/organizations. Results: Among 103,045 patients treated by 25,736 unique providers and 11,163 organizations, the percent receiving early PC increased from 0.98% in 2010 to 10.64% in 2019. Although statistically significant increases were observed across cancer types, receipt in patients with prostate cancer was relatively lower compared to pancreas, SCL, NSCL, and breast cancers (Table). After adjusting for patients’ characteristics, variation in early PC use between treating providers and organizations explained 47.9% and 31.2% of the total variation, respectively. Conclusions: Despite considerable growth in early PC receipt, utilization remained low in 2019. The large variation between providers and organizations suggests important modifiable provider behaviors and organizational characteristics in early PC use, which warrant future research. [Table: see text]

IAM1363-01: A phase 1/1b study of a selective and brain-penetrant HER2 inhibitor for HER2-driven solid tumors.

TPS3186 Background: HER2 activation promotes carcinogenesis, prompting the development of HER2-directed therapies in cancers with HER2 alterations. There is a critical need for novel agents to treat HER2-driven cancers, as current inhibitors display significant toxicity and lack efficacy against certain mutations. IAM1363 (formerly IAM-H1 and ENT-H1) is an irreversible and brain-penetrant small molecule inhibitor targeting both wildtype HER2 and oncogenic HER2 mutants, including the recalcitrant exon 20 mutations. IAM1363, discovered through Iambic Therapeutics' AI-driven discovery platform, represents the only example of a Type II HER2 inhibitor. It was designed to avoid off-target toxicities from EGFR inhibition thus expanding the therapeutic index over existing HER2 inhibitors. In preclinical studies, IAM1363 shows preferential tumor enrichment, a unique feature observed across tumor types and anatomic sites. The promising target specificity and pharmacokinetic profile of IAM1363 offers the potential to bridge gaps left by current standard of care HER2 targeted treatments. Methods: IAM1363-01 is a first-in-human, multicenter, open-label, Phase 1/1b dose escalation and expansion trial evaluating IAM1363 as monotherapy and in combination with trastuzumab. The study will enroll patients with relapsed/refractory malignancies that have documented HER2 gene alterations (amplification or tyrosine kinase domain mutations) or protein overexpression. Prior treatment with HER2 directed therapies is allowed after a washout period. Part 1, dose escalation, will enroll patients with any tumor type. Part 2 will optimize IAM1363 dosing as monotherapy and in combination with trastuzumab. Part 3 consists of a Simon 2-Stage evaluation of IAM1363 as monotherapy in patients with colorectal, non-small cell lung, and bladder cancer with or without brain metastases and in combination with trastuzumab in breast cancer patients with brain metastases. Key objectives are to establish the safety and tolerability of IAM1363, characterize PK/PD, and determine the recommended Phase 2 dose. Exploratory analysis will correlate HER2 expression level/mutation status with tumor response and tolerability. IAM1363 will be given orally once daily at a starting dose of 120 mg/day in up to 4 planned dose levels. Dose escalation employs an accelerated titration design that transitions to a standard 3 + 3 design based on protocol-specific criteria. Up to 287 patients will be enrolled at approximately 20 sites and treated until disease progression or unacceptable toxicity. The trial is designed to establish the safety profile of IAM1363 as a single agent and in combination with trastuzumab and to provide initial proof of antitumor activity in malignancies with HER2 overexpression, gene amplification, or tyrosine kinase mutations, including in patients with brain metastases. Clinical trial information: NCT06253871 .

Enrollment of older adults in clinical trials for oncology indications, 2010-2020: An evaluation by the FDA.

e23029 Background: Underrepresentation of older adults (≥65 years) in clinical trials may limit important evidence for regulatory and clinical decision-making. Prior FDA analysis1 demonstrated underrepresentation of older adults in oncology clinical trials. With recent efforts to encourage inclusion of older adults, we evaluated representation in clinical trials using updated data. Methods: This analysis used patient-level data from pivotal trials supporting approvals of new drugs and biologics, and expanded indications from 2010-2020 for cancers with high incidence in the U.S. population and older adults: breast (BC), colorectal (CRC), non-small cell lung (NSCLC), prostate (PC), chronic myeloid leukemia (CML), and multiple myeloma (MM). Age distribution of trial participants compared to incident cases from the NCI SEER Program was analyzed using the Enrollment to Incidence Ratio (EIR). For each cancer-age group, EIR is the percentage of trial participants divided by percentage of incident cases. Previously established EIR categories of < 0.8, 0.80-1.20, and > 1.2 indicate under-, adequate, and over-representation, respectively. Results: The analysis included 86000 participants in 152 trials. Adults >65 yrs were underrepresented in BC trials. Adults >75 yrs were underrepresented in BC, CRC, NSCLC, CML, and MM trials (Table 1). Conclusions: Despite slight variation across cancers, adults >75 yrs appeared to be underrepresented in clinical trials, excepting PC. Given heterogeneity in the natural history by cancer site, incidence may not best represent age distribution of the intended use population for all trials. Further evaluation using prevalence as a comparator may provide additional insights. Additionally, results may be affected by clinical factors such as stage; however, comparison is limited by SEER data element availability. Planned sensitivity analyses such as stratification by drug class and line of therapy may further elucidate enrollment patterns.

EVEREST-1: A seamless phase 1/2 study of A2B530, a carcinoembryonic antigen (CEA) logic-gated Tmod CAR T-cell therapy, in patients with solid tumors associated with CEA expression also exhibiting human leukocyte antigen (HLA)-A*02 loss of heterozygosity (LOH).

TPS2698 Background: Chimeric antigen receptor (CAR) T-cell therapies for treating solid tumors are challenging due to a lack of tumor-specific targets that discriminate cancer from normal cells. Previous studies using CEA T-cell receptors and T-cell engagers have resulted in dose-limiting, on-target, off-tumor toxicities (1,2). A2B530 is an autologous logic-gated, CEA-targeted Tmod CAR T-cell therapy that addresses the challenges of on-target, off-tumor toxicity by combining a CAR-activating receptor with a blocking receptor to discriminate tumor from normal cells (3). The activator recognizes CEA on the surface of both tumor and normal cells while specificity for tumor cells is provided by a blocker that binds HLA-A*02. In patients with both germline HLA-A*02 and tumor-associated HLA-A*02 LOH, the blocker prevents on-target, off-tumor toxicity on normal cells owing to retained HLA-A*02 expression (4). HLA-A*02 LOH can be detected using next-generation sequencing (Tempus AI, Inc.). With this definitive discriminator target, A2B530 can potentially provide a therapeutic window to treat patients with CEA-expressing solid tumors exhibiting HLA LOH. Methods: EVEREST-1 (NCT05736731) is a seamless, phase 1/2, open-label, nonrandomized study to evaluate the safety and efficacy of A2B530 in adult patients. Patients are enrolled through BASECAMP-1 (NCT04981119), a master prescreening study that identifies patients with HLA LOH at any time in the course of their disease. BASECAMP-1–eligible patients undergo leukapheresis and, when clinically appropriate, their banked T cells are manufactured for the EVEREST-1 study. The key inclusion criteria include histologically confirmed recurrent unresectable, locally advanced, or metastatic cancers associated with CEA expression: non-small cell lung, colorectal, or pancreatic cancers. Patients should have received ≥1 line of prior therapy such as checkpoint inhibitor, molecular targeted, or chemotherapy. The primary objective of phase 1 is to evaluate the safety and tolerability of A2B530 and determine the recommended phase 2 dose (RP2D). The dose-escalation portion is based on the Bayesian optimal interval design. The dose-expansion phase will confirm RP2D and collect biomarker data to further characterize A2B530. Phase 2 will assess overall response rate per RECIST v1.1. As of January 29, 2024, 8 patients have been enrolled on EVEREST-1. A2B530 was successfully manufactured for all patients, and all patients have received A2B530 infusion, with the first patient dosed in May 2023. Dose escalation is ongoing. 1. Parkhurst, et al. Mol Ther. 2022. 2. Tabernero, et al. J Clin Oncol. 2017. 3. Hamburger, et al. Mol Immunol. 2020. 4. Sandberg, et al. Sci Transl Med. 2022. Clinical trial information: NCT05736731 .

Shareable artificial intelligence to extract cancer outcomes from electronic health records.

11000 Background: Clinical outcomes such as response, progression, and metastasis represent crucial data for observational cancer research, but outside of clinical trials, such outcomes are usually recorded only in unstructured notes in electronic health records (EHRs). Manual EHR annotation is too resource-intensive to scale to large datasets. Individual cancer centers have trained artificial intelligence natural language processing (AI/NLP) models to extract outcomes from their EHRs. However, due to concerns that models trained on protected health information (PHI) might encode private data, such models usually cannot be exported to other centers. Methods: EHR data from Dana-Farber Cancer Institute (DFCI) and Memorial-Sloan Kettering (MSK) collected through the AACR Project GENIE Biopharma Collaborative were used to train and evaluate Bidirectional Encoder Representations from Transformers (BERT)-based NLP models to extract outcomes from imaging reports and oncologist notes annotated with the Pathology, Radiology/Imaging, Signs/Symptoms, Medical oncologist, and bioMarkers (PRISSMM) framework. Document-level outcomes included the presence of active cancer; response; progression; and metastatic sites. ‘Teacher’ models trained on DFCI EHR data were used to label imaging reports and discharge summaries from the public MIMIC-IV dataset. ‘Student’ models trained to use MIMIC documents to predict teacher labels were transferred to MSK for evaluation. Results: Teacher models were trained at DFCI on 30,332 imaging reports for 2609 patients and 32,173 oncologist notes for 2917 patients with non-small cell lung, colorectal, breast, prostate, pancreatic, or urothelial cancer. The models were used to label 217,642 imaging reports and 141,377 discharge summaries from MIMIC for student model training. These DFCI-trained student models were evaluated at MSK, demonstrating high discrimination (AUROC > 0.90) across outcomes. Conclusions: This privacy-preserving “teacher-student” AI/NLP framework could expedite linkage of genomic data to clinical outcomes across institutions, accelerating precision cancer research. [Table: see text]

Human-AI teams to improve accuracy and timeliness of oncology trial prescreening: Preplanned interim analysis of a randomized trial.

1524 Background: Identifying eligible patients for oncology clinical trials (“prescreening”) relies on manual chart review by clinical research coordinators (CRCs), which is time-consuming and often inaccurate. Consequently, 70% of patients with cancer who meet trial eligibility criteria are not offered participation. Natural language processing (NLP) may improve the accuracy and timeliness of prescreening. Methods: This was a preplanned interim analysis of a paired-design noninferiority trial, powered to assess timeliness. We adapted NLP algorithms to identify 13 common eligibility criteria related to cancer type/stage, prior systemic therapy, actionable biomarkers, and response criteria. NLP systems performed optical character recognition, entity and relationship extraction, and natural language inference through deep learning and symbolic AI techniques. Deidentified unstructured electronic health records (EHRs) from real-world patients with non-small cell lung (NSCLC) or colorectal (CrCa) cancer were presented to CRCs via a secure platform. Two CRCs were randomized 1:1 to view blocks of 20 charts each with (Human+AI) or without (Human-alone) NLP annotations. In pre-trial assessment, the CRCs had 86% inter-rater agreement. The primary outcome was overall chart-level accuracy, defined as percent of the 13 CRC-coded eligibility items matching a gold standard set, determined by 2-3 clinicians blinded to experimental arms. Paired t-tests compared overall accuracy (noninferiority margin 5%) and criteria-specific accuracy. Significant differences were assessed for superiority using two-sided paired t-tests. Mann-Whitney tests compared the secondary outcome of timeliness (time per chart review). Results: Among 74 (40 NSCLC; 34 CrCa) of a planned 400 patients, overall accuracy for Human+AI was noninferior to Human-alone (78.7% vs. 76.7%, mean difference 2.0%, p<0.001 rejecting inferiority); both were greater than AI-alone (63.5%). Superiority for Human+AI was demonstrated for RECIST response, but not for overall nor other criteria-specific accuracy (Table). Median time per review was lower for Human+AI than Human-alone (34.1 vs 43.9 min, adjusted p=0.05). Conclusions: Human+AI teams can improve timeliness of trial prescreening with noninferior accuracy. This platform is being used for eligibility assessment in an ongoing clinical trial. [Table: see text]

Patient-reported experience with an immunotherapy telehealth platform.

1637 Background: The Making Telehealth Delivery of Cancer Care at Home Effective and Safe for Immunotherapy (MATCHES-IO) intervention seeks to improve the efficiency and patient experience for those treated with single agent pembrolizumab. Because pembrolizumab is administered as an outpatient infusion every 3-weeks, patients require up to 18 clinic visits per year to receive treatment, which is arduous. During the COVID-19 pandemic, the FDA granted accelerated approval for an extended interval dosing administered every 6 weeks, but despite this approval every 3-week dosing remains the standard (65% of prescriptions) as clinicians perceive this schedule enables them to identify and manage toxicity early. Telehealth may be the impetus to change the standard to the more convenient 6-week in person schedule, but evidence is needed. Methods: We conducted a single arm pragmatic trial to evaluate the efficiency and patient experience of a telehealth immunotherapy platform (MATCHES-IO) in patients with non-small cell lung, genitourinary, or melanoma cancers receiving single agent pembrolizumab. MATCHES-IO evaluates whether in-person visits for pembrolizumab therapy q6wk (rather than q3wk), with interim telehealth toxicity checks between in-person treatments, is more efficient and enhances patient experience relative to the standard q3wk infusion visits. The components of the platform include clinician-patient virtual visits, labs at home, biometric devices at home for vital sign monitoring, and electronic patient-reported outcomes to monitor for common IO-related toxicities. Patient experience was assessed after each MATCHES-IO televisit for up to two televisits. We measured experience with a patient experience survey that included the net promoter score, of how likely are they to recommend this intervention to similar patients (scale 0-10, higher score is desirable). A score of 9 or greater is considered excellent. Results: Between July 2023 and February 2024, 27 patients were enrolled, median age 68 (range 45–85), 74% white, 11% black, 11% Asian, and 41% female. Cancer types included thoracic (59% of patients), genitourinary (33%), and melanoma (7%). 19 patients (70.4%) have completed a MATCHES-IO televisit and completed the patient experience survey. The median net promoter score was 9 (range 2-10). 94% patients perceived a benefit to the MATCH-IO televisit including saved time (83% of respondents), patient convenience (83%), convenience for caregiver/family (44%), saved money (44%), and better monitoring of cancer and treatment (39%). 78% of patients found the at home visit less stressful than the in-person visit. Conclusions: Patients endorsed an enhanced experience with an immunotherapy telehealth platform for extended dosing of pembrolizumab. Further follow-up is needed to confirm these experience findings and determine whether this platform improved efficiency through fewer in-person visits.

A phase 1/2, first-in-human study of DS-3939a in patients with advanced solid tumors: A new DXd ADC targeting TA-MUC1.

TPS3165 Background: Mucin 1 (MUC1), a single-transmembrane glycoprotein, is expressed on the apical membrane of the epithelial surface. In normal tissues, it is highly glycosylated and protects the underlying epithelia. In cancer cells, however, MUC1 is hypoglycosylated due to changes in the expression patterns of some sialyltransferases, and exposes new epitopes on MUC1 in tumors, referred to as tumor-associated MUC1 (TA-MUC1). TA-MUC1 loses cell polarity and is redistributed over the cell surface and within the cytoplasm. Overexpression of TA-MUC1 in several malignancies is associated with poor prognosis and development of metastasis, thus making it an attractive therapeutic target. DS-3939a is a novel antibody-drug conjugate in development for the treatment of malignant tumors composed of a humanized anti-TA-MUC1 antibody, a peptide-based cleavable linker, and a potent topoisomerase I inhibitor (DXd). This study will assess the safety, tolerability, and efficacy of DS-3939a in patients with advanced solid tumors. Methods: This phase 1/2, first-in-human, open-label, multicenter, 2-part, dose-escalation, and dose-expansion study (NCT05875168) is active and plans to enroll up to 430 adult patients. Part 1 (dose-escalation) is accruing patients with locally advanced, metastatic, or unresectable urothelial, non-small cell lung, breast, ovarian, biliary tract, or pancreatic cancers, and Part 2 (dose-expansion) will enroll patients with various advanced solid TA-MUC1-expressing tumors. In Parts 1 and 2, DS-3939a will be administered intravenously on Day 1 of a 21-day cycle. The primary endpoints in Parts 1 and 2 include safety and tolerability of DS-3939a as assessed by the number of patients with dose-limiting toxicities (Part 1 only), treatment-emergent adverse events, and other safety parameters. Part 2 will also evaluate efficacy by objective tumor response rate per RECIST version 1.1 as a primary endpoint. Secondary endpoints include the disease control rate, duration of response, time to response, progression-free survival, overall survival, TA-MUC1 expression (detected by immunohistochemistry), the pharmacokinetic profile of DS-3939a, and the number of patients with anti-drug antibodies against DS-3939a. Clinical trial information: NCT05875168 .