Citations

e20700 Background: Leptomeningeal metastases (LM) are a particularly challenging complication of EGFR-mutant non–small cell lung cancer (NSCLC) and are often associated with rapid neurologic decline. Intrathecal (IT) chemotherapy is frequently used in this setting, but its efficacy and safety have not been well defined. We conducted a systematic review and meta-analysis to assess neurologic response and treatment-related toxicity associated with IT chemotherapy in EGFR-mutant NSCLC with LM. Methods: We performed a PRISMA-guided search of MEDLINE, EMBASE, Cochrane CENTRAL, and targeted grey literature to identify prospective and retrospective studies reporting outcomes of IT chemotherapy in EGFR-mutant NSCLC with LM. Quantitative analyses were restricted to studies with extractable endpoints. Neurologic response and grade ≥3 toxicity were pooled using random-effects generalized linear mixed models with logit transformation. Heterogeneity was assessed using I² and τ². Prespecified sensitivity analyses examined neurologic response by route of administration (lumbar puncture vs Ommaya reservoir), concomitant EGFR-TKI use, and study design or era. Results: Four studies comprising 76 patients reported extractable neurologic response data. The pooled neurologic response rate was 81.9% (95% CI, 64.0%–92.0%), with moderate heterogeneity (I² = 63.4%, p = 0.04). Grade ≥3 toxicity was reported in three studies (n = 59), with a pooled incidence of 11.8% (95% CI, 3.2%–34.9%) and substantial heterogeneity (I² = 71.3%, p = 0.03). Myelosuppression was the most frequently reported severe adverse event, and no consistent signal of severe neurologic toxicity was identified. Exploratory sensitivity analysis suggested higher neurologic response in LP-treated cohorts compared with mixed LP/Ommaya cohorts; however, this finding was driven by a single small study and is likely influenced by limited sample size and selection bias and should be considered hypothesis-generating only. Neurologic response was otherwise consistent across study designs and eras. Stratification by concomitant EGFR-TKI use was limited by incomplete reporting. Median overall survival was variably reported and ranged from approximately 7 to 12 months. Conclusions: In EGFR-mutant NSCLC with leptomeningeal metastases, IT chemotherapy is associated with a high pooled neurologic response and relatively uncommon severe toxicity, although heterogeneity across studies remains substantial. The observed neurologic benefit appears consistent across study designs and is more likely related to underlying EGFR-sensitive disease biology than to a definitive effect of route of administration. These findings support further prospective evaluation of IT chemotherapy as part of multimodal treatment strategies in this high-risk population.

e20648 Background: Immune checkpoint inhibitors have become a mainstay of treatment for lung cancer. However, their effectiveness varies, and is strongly influenced by patient characteristics. Smoking is the most important etiologic factor for lung cancer, and is expected to be associated with a high tumor mutational burden, which may in turn influence immunotherapy response. However, the extent to which smoking status modifies the benefit of immunotherapy in real-world populations is not well established. Methods: We identified 43,437 cases of invasive lung cancer with documentation of smoking status from the California Cancer Registry 2010-2021. Immunotherapy was documented as given or not given. Smoking status was defined as never, current or former. Cox regression was used to assess the unadjusted risk of mortality associated with immunotherapy and after adjusting for age, sex, race/ethnicity, stage, socio-economic status (SES), Charlson Comorbidity Index and insurance status. Analyses were conducted separately for current and former and never smokers. Results: Smoking status was documented for 43,347 cases (never = 15.887, current = 27,056, and former = 404). There were 4,014 cases treated with immunotherapy. The unadjusted risk of mortality for patients treated with immunotherapy was increased in non-smokers whereas immunotherapy reduced the risk of mortality in current and former smokers. After adjustment, all patients had a reduced risk of mortality, but current and former smokers had a higher risk reduction than non-smokers. Conclusions: Smoking modifies the effect of immunotherapy in patients with lung cancer, and our data suggests patients who currently or formerly smoked may derive a greater benefit from immunotherapy. This will need careful patient messaging as we continue public efforts towards smoking cessation. Unadjusted and adjusted* hazard ratios and 95% confidence intervals for immunotherapy. Unadjusted Adjusted* HR (95%CI) HR (95%CI) Never smokers 1.35 (1.26, 1.45) 0.76 (0.70, 0.82) Current or former smoker 0.86 (0.82, 0.91) 0.52 (0.49, 0.55) *Adjusting for age, sex, race/ethnicity, stage, SES, Charlson Comorbidity Index and insurance status.

e20629 Background: Concomitant inhibition of the vascular endothelial growth factor (VEGF) pathway together with PD-1 or PD-L1 immune checkpoint blockade has demonstrated synergistic activity in non-small cell lung cancer (NSCLC). However, comparative evidence across individual immune checkpoint inhibitor (ICI) and anti-VEGF combinations remains limited. This network meta-analysis (NMA) aimed to evaluate and compare the efficacy of available ICI plus anti-VEGF combination strategies in NSCLC. Methods: A systematic search of three electronic databases and two clinical trial registries was conducted to identify randomized controlled trials evaluating combinations of ICIs with anti-VEGF agents. A frequentist NMA framework was applied using the netmeta package in R. Hazard ratios (HRs) were used for overall survival (OS) and progression-free survival (PFS), while odds ratios (ORs) were used for objective response rate (ORR). Treatment ranking was performed using P scores. Results: A total of 24 randomized controlled trials were included in the network. For survival outcomes, the combination of nivolumab plus bevacizumab demonstrated significantly improved OS (HR = 0.66, 95% CI 0.50–0.88) and PFS (HR = 0.46, 95% CI 0.27–0.81) compared with standard chemotherapy. In addition, atezolizumab plus bevacizumab was associated with a significant OS benefit over chemotherapy (HR = 0.79, 95% CI 0.65–0.95). Indirect comparisons showed that nivolumab plus bevacizumab significantly improved PFS compared with pembrolizumab monotherapy, as well as pembrolizumab combined with lenvatinib or chemotherapy. Furthermore, nivolumab combined with sitravatinib demonstrated significantly superior OS and PFS compared with ramucirumab and chemotherapy. For ORR assessed using RECIST 1.1, nivolumab-based combinations could not be incorporated into the network due to insufficient data, and no other combination demonstrated a statistically significant advantage over chemotherapy. However, indirect comparisons showed that pembrolizumab plus lenvatinib was superior to lenvatinib monotherapy (OR = 4.76, 95% CI 1.04–25.0). P-score rankings identified nivolumab plus bevacizumab as the top-ranked regimen for survival outcomes, followed by atezolizumab plus bevacizumab. For ORR, pembrolizumab plus lenvatinib ranked highest, followed by atezolizumab plus bevacizumab. Conclusions: Nivolumab-based ICI and anti-VEGF combinations, particularly with bevacizumab and sitravatinib, demonstrate the most favorable survival outcomes in NSCLC. Evidence for ORR remains limited, and further randomized comparisons are required to better define the relative efficacy of these combinations.

e20595 Background: Circadian regulation of immune function has been proposed as a modifier of immune checkpoint inhibitor (ICI) efficacy, with retrospective studies suggesting improved outcomes with earlier infusion timing. Whether these findings translate to routine oncology practice, where ICIs are administered across multiple lines, frequently combined with chemotherapy, and subject to treatment interruptions, remains uncertain. We evaluated the association between ICI infusion timing and overall survival (OS) in a real-world cohort of patients with non-small cell lung cancer (NSCLC). Methods: A retrospective cohort study included adults with NSCLC who received at least one line of ICI therapy at a community oncology center between January 2018 and December 2024. Infusion timing was defined at the patient level as morning-only ( < 1300 hours), evening-only (≥1300 hours), or mixed timing, reflecting receipt of infusions exclusively in different time-of-day categories across treatment lines. OS was measured from diagnosis to death or last follow-up. Kaplan-Meier and Cox proportional hazards models were used. Multivariable analyses adjusted for stage, ECOG performance status, body mass index, and premature ICI discontinuation. Line of therapy was excluded due to collinearity with mixed-timing classification and risk of immortal time bias. Results: A total of 127 patients were included (morning-only n = 77; mixed timing n = 36; evening-only n = 14). Staging data were available for 99.2%, and 94 deaths occurred. Median OS was 495 days for morning-only, 850 days for mixed timing, and 388 days for evening-only patients. Kaplan-Meier analysis showed no statistically significant OS differences across groups. In sensitivity analysis excluding patients with mixed infusion timing, morning-only and evening-only groups demonstrated overlapping survival curves, with no evidence of a time-of-day–specific survival advantage. In univariate analysis, mixed timing was associated with a borderline reduction in mortality compared with morning-only infusions (HR 0.63, 95% CI 0.39-1.00), while evening-only timing was not (HR 1.63, 95% CI 0.85-3.16). In multivariate analysis restricted to stage III-IV disease, infusion timing was not independently associated with OS. ECOG performance status, premature ICI discontinuation and advanced staging were the strongest predictors of mortality. Conclusions: In this real-world NSCLC cohort, ICI infusion timing was not independently associated with overall survival after adjustment for patient fitness and treatment continuity. These findings suggest that in pragmatic settings, clinical factors may outweigh subtle chronotherapy effects reported in more selected populations.

e20511 Background: Trop-2 ADCs have emerged as a novel therapeutic approach in NSCLC that is refractory to chemotherapy and immune checkpoint inhibitors. Comparative interpretation of the efficacy of multiple Trop-2 ADCs, including Sacituzumab tirumotecan (Sac-TMT), Datopotamab deruxtecan (Dato-DXd), and Sacituzumab govitecan (SG), is limited. We therefore conducted a systematic review, IPD survival analysis, and meta-analysis to evaluate the efficacy and PROs of Trop-2 ADCs compared with chemotherapy in NSCLC. Methods: A systematic search was conducted from inception to October 31, 2025, across PubMed, Scopus, Cochrane, and CT.gov. Clinical trials evaluating Overall Survival (OS) and Progression-free survival (PFS) in patients treated with Trop-2 ADC alone or compared with Chemotherapy were included. IPD were extracted from the Kaplan-Meier curves of clinical trials for reconstruction using the KMfromIPD website by MD Anderson, which stratified the data by treatment group. Outcomes included OS, PFS, and time to deterioration (TTD). The cumulative data were analyzed using R 4.5.1 and RStudio. RevMan version 5.4.1 was used for the meta-analysis of subgroups. Heterogeneity was assessed using the I 2 statistic. Results: Of 513 articles, 7 phase 1/2 single-arm and 4 randomized control trials were included, comprising 2,619 patients.Trop-2 ADCs were associated with improved OS (Median 15.6 vs 12.5 months; HR 0.78, 95% CI 0.69-0.88; p < 0.0001) and PFS (Median 5.5 vs 4.1 months; HR 0.66, 95% CI 0.59-0.74; p < 0.0001). Sac-TMT demonstrated the greatest OS benefit (OS 23.3 months; HR 0.40). Further drug classes are depicted in Table 1. PROs assessment showed TROP-2 ADCs were associated with longer time to deterioration in global quality of life and dyspnea-free survival compared to chemotherapy (HR 0.8 and 0.69), (p = 0.04 and p < 0.0001, respectively). Exploratory subgroup analysis showed improved OS among patients with actionable gene alteration (AGA) treated with Trop-2 ADCs (HR 0.58, p < 0.00001). However, no significant difference was observed in either treatment arm among patients without AGA (HR 0.89, p = 0.15). Conclusions: Trop-2 ADCs were associated with improved survival and PROs compared with chemotherapy in NSCLC, with variability across agents. Sac-TMT has shown higher OS and PFS. Dato Dxd had a favorable PFS; however, the OS was not superior to chemo. Furthermore, SG showed no favorable clinical outcomes compared with the chemo group. These findings support the continued development of Trop-2-directed therapies and highlight the need for prospective comparative trials and biomarker-driven patient selection. Drug N mOS (months) HR p- value mPFS (months) HR p- value Sac-TMT 386 23.32 0.4 <0.000001 7.5 0.5 <0.000001 Dato-DXd 486 13.2 0.95 0.55 5.02 0.71 <0.000001 SG 347 12.61 1.02 0.82 4.18 0.87 0.05

e20139 Background: Extensive-stage small cell lung cancer (ES-SCLC) frequently presents with brain metastases, affecting approximately 20% of the patients at diagnosis. The prognostic impact of baseline brain metastases on survival outcomes in patients treated with contemporary chemo-immunotherapy remains unclear, creating uncertainty in first-line treatment selection for this high-risk population. We conducted a systematic review and meta-analysis to evaluate the impact of baseline brain metastases on treatment outcomes in ES-SCLC. Methods: This PRISMA-compliant systematic review and meta-analysis was prospectively registered in PROSPERO. PubMed, Embase, Cochrane CENTRAL, and ClinicalTrials.gov were systematically searched from database inception through January 2026 to identify randomized controlled trials evaluating PD-1/PD-L1-based chemo-immunotherapy versus platinum-etoposide chemotherapy in adults with ES-SCLC. Two reviewers independently screened studies and extracted data, with discrepancies resolved by consensus. Hazard ratios (HRs) for overall survival (OS) and progression-free survival (PFS) were pooled separately for patients with baseline brain metastases (BM+) and without brain metastases (BM−) using random-effects models with Hartung–Knapp adjustment. Analyses were performed using R. Prespecified subgroup analyses were conducted based on follow-up duration and immunotherapy regimen. Results: Twelve randomized controlled trials were included. Among patients with baseline brain metastases (BM+), pooled OS from seven studies did not show a significant survival benefit (HR 0.84, 95% CI 0.66–1.08, I² = 0.0%, P = 0.145), nor did pooled PFS from five studies (HR 1.03, 95% CI 0.60–1.79, I² = 0.0%). Subgroup analyses revealed no significant heterogeneity. In contrast, among patients without baseline brain metastases (BM−), pooled OS from nine studies demonstrated a significant survival benefit (HR 0.72, 95% CI 0.69–0.76, I² = 0.0%, P < 0.001), with corresponding improvement in PFS (HR 0.70, 95% CI 0.60–0.82, I² = 0.0%). These findings indicate a differential treatment effect according to baseline brain metastasis status. Conclusions: First-line chemo-immunotherapy does not apprear to confer a significant survival benefit in ES-SCLC patients with baseline brain metastases, reflecting a differential treatment effect by brain metastasis status. These findings suggest that baseline brain metastases should be considered in treatment selection and future clinical trial design and highlight the need for optimized systemic strategies for this high-risk population.

e20030 Background: Recent studies suggested associations between higher serum Magnesium (Mg) level and better survival outcome in patients receiving immunotherapy, especially in small cell lung cancer (SCLC). However, the prognostic significance of Mg in extensive-stage (ES) SCLC treated with first-line chemoimmunotherapy remains unclear. We evaluated the association between baseline serum Mg and survival outcomes in ES-SCLC receiving platinum–etoposide plus PD-(L)1 blockade. Methods: We conducted a retrospective cohort study of adults with ES-SCLC treated with first-line platinum–etoposide plus atezolizumab or durvalumab at Mayo Clinic via Mayo Clinic Platform in 2018–2025. Baseline average serum Mg level was defined as the mean of all Mg labs from day −30 to day 0 relative to chemoimmunotherapy initiation and categorized as low ( < 1.7 mg/dL), normal (1.7-2.0 mg/dL), or high ( > 2.0 mg/dL) using clinical cutoff. Low and normal groups were combined a priori due to small low-Mg counts. The primary endpoint was overall survival (OS), estimated by KM and compared by log-rank test. OS was measured from chemoimmunotherapy initiation to death and censored at last follow-up. Multivariable Cox proportional hazards models adjusted for age, gender, ECOG, liver/brain metastases status, and smoking status. Results: The cohort included 214 patients, 124 (58%) were female; median age was 68 years; 49% had liver metastases and 30% had brain metastases. Median OS was 10.4 months (95% CI 7.6-15.9) among 99 patients with high serum Mg level, compared with 7.5 months (95% CI 6.7-10.7) in 115 patients with low or normal serum Mg level (Log-rank p = 0.126). High Mg was associated with a trend toward longer OS in unadjusted analyses (Hazard ratio [HR] 0.80, 95% CI 0.59-1.07, p = 0.129); however, the association was not statistically significant after multivariable adjustment (aHR 0.81, 95% CI 0.60-1.10, p = 0.180). Conclusions: Among patients with ES-SCLC treated with first-line platinum–etoposide plus PD-(L)1 blockade, higher baseline serum Mg was associated with numerically longer OS but was not independently associated with survival after multivariable adjustment. These findings suggest baseline serum Mg level may be associated with prognosis and warrant validation in larger cohorts. Baseline characteristics by baseline serum magnesium category in ES-SCLC treated with first-line chemoimmunotherapy. Variable Overall (N=214) High Mg >2.0 (n=99) Low/Normal Mg ≤2.0 (n=115) p value* Age, median (IQR) 68 (62, 74) 68 (61, 74) 69 (62, 75) 0.201 Female, n (%) 124 (58%) 57 (58%) 67 (58%) >0.999 ECOG ≥2, n (%) 58 (27.1%) 28 (28%) 30 (26%) 0.719 Liver metastases, n (%) 105 (49%) 49 (49%) 56 (49%) 0.648 Brain metastases, n (%) 65 (30%) 33 (33%) 32 (28%) 0.469 Ever smoker, n (%) 197 (92.1%) 90 (90.9%) 107 (93.0%) 0.974 *p values: Wilcoxon (age); χ²/Fisher (categorical).

e20000 Background: Multimodal machine-learning (ML) models that fuse imaging, pathology, omics, and clinical data may improve overall-survival (OS) prediction in non–small-cell lung cancer (NSCLC) beyond staging-based tools. We systematically reviewed the design, performance, and methodological quality of these models. Methods: Following PRISMA 2020, we searched Ovid (MEDLINE/Embase/CENTRAL/CDSR), Scopus, IEEE Xplore, and arXiv (January 2017–July 2025). Eligible studies developed or validated ML models integrating ≥2 modalities to predict OS in adults with NSCLC and reported either time-to-event or fixed-horizon binary outcomes. Two reviewers independently screened, extracted, and assessed risk of bias using PROBAST with PROBAST-AI items. Due to heterogeneity, results were synthesized narratively. Results: We included 18 studies (2021–2025) with per-study sample sizes ranging from 115 to 2,898. Outcome framing: time-to-event OS only (n=11), fixed-horizon binary OS only (n=4), and both (n=3). Modalities most often used were clinical structured data (15/18), CT (12/18), PET (6/18), molecular omics (7/18), pathology whole-slide images (4/18), and EHR text (1/18). Fusion strategies clustered as early/concatenation (10/18), interaction-based (attention/bilinear/graph; 5/18), and late/score-level (3/18). For time-to-event OS, internal C-indices ranged 0.658–0.893, with the highest internal value 0.893 (CT+clinical). One study reported external C-index (0.678, pathology+genes). An additional study reported external time-dependent AUC 0.845 at 1-year for a PET/CT-genomics survival model (n=32). For binary OS, internal AUROCs were 0.802–0.888 (2–5-year horizons), and internal accuracies ranged 0.68–0.93 (1–5 years). External binary performance included accuracy 0.72 at 1-year in an immunotherapy cohort. Across studies, multimodal models typically outperformed the best single-modality comparator by ~+0.06 C-index or AUROC, though absolute gains varied. Risk of bias was frequently high in the analysis domain (internal-only validation, optimistic tuning, sparse calibration reporting); code/weights were publicly available in 5/18 studies. Conclusions: Multimodal ML models for NSCLC show consistent, modest improvements in OS discrimination versus single-modality approaches, with CT+clinical the most translationally pairing. However, independent validation, calibration, and transparency remain limited, constraining clinical adoption. Future work should prioritize multi-center datasets, standardized reporting, and open workflows.

e19564 Background: Risk stratification in multiple myeloma (MM) relies on baseline staging systems and cytogenetic abnormalities, yet early relapse occurs in a subset of patients despite favorable risk features. Prior studies showed that clonal plasma cell S-phase assessed by flow cytometry using the plasma cell proliferation (PCPRO) assay at diagnosis and at autologous stem cell transplantation (ASCT) identifies biologically high-risk disease. The prognostic significance of clonal plasma cell S-phase following ASCT has not been well defined. Methods: We retrospectively analyzed MM patients who underwent ASCT within one year of diagnosis between January 1, 2013, and August 31, 2024. Patients without available post-ASCT clonal plasma cell S-phase assessment were excluded, yielding 150 evaluable patients. The proportion of clonal plasma cells in S-phase was determined by DNA content between G0/G1 and G2/M peaks and reported as a percentage. S-phase was assessed at the first post-ASCT marrow evaluation at day +60 (D45–75) and/or day +100 (D76–130), using the earlier assessment when both were available. Patients were classified as low (<2%) or high (≥2%) post-ASCT S-phase. Landmark progression-free survival (PFS) analyses were performed from the post-transplant assessment. Multivariable Cox models adjusted for cytogenetic risk, maintenance therapy, disease response at transplantation, International Staging System (ISS), and age. Results: Among 150 evaluable patients, 36 (24%) had high post-ASCT S-phase (≥2%) and 114 (76%) had low post-ASCT S-phase (<2%). Median age at ASCT was higher in the high S-phase group (64.5 vs 61.0 years; p=0.063), and advanced ISS stage (II/III) was more frequent (78% vs 57%; p=0.02). Cytogenetic risk, induction regimen, depth of response at transplantation, and maintenance therapy were similar between groups. Patients with high post-ASCT S-phase had significantly inferior PFS compared with those with low S-phase (median 17.0 vs 39.7 months; p=0.0003). On multivariable analysis, post-ASCT S-phase ≥2% remained independently associated with inferior PFS (HR 1.76, 95% CI 1.00–3.09; p=0.049). In an exploratory analysis among patients with paired S-phase assessments at the time of ASCT and post-ASCT, dynamic S-phase trajectories further stratified outcomes. Median PFS was longest in patients with persistently low S-phase (low→low; 39.7 months), intermediate in those converting from high to low S-phase (high→low; 27.2 months), and shortest in patients with newly emergent or persistently high post-ASCT S-phase (low→high: 15.0 months; high→high: 11.0 months) p=0.0019. Conclusions: Elevated proliferative activity of residual clonal plasma cells following ASCT identifies a biologically high-risk subset of MM patients with early relapse. Dynamic changes in post-ASCT S-phase further refine post-transplant risk stratification.

e19562 Background: Multiple myeloma is a chronic blood cancer in which patients often live for many years, yet the risk of death remains ongoing. While survival has improved over time, population-level studies that combine traditional survival analysis, advanced modeling techniques, and future burden projections are limited. Methods: We performed a population-based retrospective study using the Surveillance, Epidemiology, and End Results (SEER) database. Overall survival was evaluated using Kaplan–Meier analysis and Cox proportional hazards regression. Parametric survival modeling was conducted using Weibull regression to better characterize long-term mortality patterns. Machine-learning survival analysis was performed using random survival forests to assess the relative importance of clinical and demographic factors. Annual case counts were analyzed and projected through 2035 using time-series forecasting methods. Results: The study included 8,627 patients with multiple myeloma. Overall survival declined gradually over long-term follow-up, with approximately one-quarter to one-third of patients surviving at extended time points. Female patients experienced better survival than males. Significant survival differences were observed across racial groups, with Asian or Pacific Islander patients demonstrating the most favorable outcomes and American Indian or Alaska Native patients the poorest. In multivariable analyses, age, sex, race, and year of diagnosis were independently associated with survival. Parametric modeling closely mirrored non-parametric survival estimates and supported a time-dependent pattern of mortality. Machine-learning analysis identified year of diagnosis as the strongest predictor of survival, followed by age and race. Forecasting suggested that the burden of multiple myeloma will remain substantial through 2035. Conclusions: Multiple myeloma is associated with long-term survival but persistent mortality risk and ongoing demographic disparities. Combining traditional survival analysis with parametric and machine-learning approaches provides a more complete understanding of outcomes and highlights the sustained future burden of this disease, with important implications for clinical care and healthcare planning.