Difference In Overall Survival Research Articles

Access to germline testing and prevalence of somatic mutations in patients with breast cancer from disadvantaged areas: A single-center study.

e13750 Background: Socioeconomic disparities affect clinical outcomes in several malignancies including breast cancer (BC). The Area Deprivation Index (ADI) is a tool used to classify regions based on socioeconomic factors, including employment, education, housing, and poverty. Here, we investigate ADI among patients (pts) with BC seen at Roswell Park Comprehensive Cancer Center and its association with recurrence free survival (RFS), time to next treatment (TNT), overall survival (OS), and access to somatic mutation via next-generation sequencing (NGS) and germline testing of BC pts. Methods: We conducted a retrospective analysis of 409 pts diagnosed with stages 1-3 and de-novo stage 4 BC from 2014 to 2018. ADI scores were derived from patients' home addresses using www.neighborhoodatlas.medicine.wisc.edu and categorized into four quartiles: Q1 (80-100%), Q2 (60-79%), Q3 (40-59%), and Q4 (0-39%) with Q1 representing areas with highest deprivation. Demographic and clinicopathological characteristics were summarized by ADI. Kruskal-Wallis and Chi-square was used for comparing continuous and categorical variables respectively. Results: Among 409 pts, 92% were female, 77% White, and 10% Black. 30% (n=123/409) lived in the most disadvantaged area Q1, 26% (106/409) Q2, 26% (104/409) Q3, and 12% (51/409) Q4, and 6% (25/409) unknown. 72% pts were hormone receptor positive, 14% HER2+, and 14% triple negative BC. No significant differences were noted in the racial distribution, employment, comorbidities, smoking, screening mammogram, type of insurance, BC stage and subtype, and adherence to therapy by ADI. Access to germline testing was significantly lower among pts from Q1 and Q2: (68/232 or 28%) vs. (65/156 or 42%) from Q3 and Q4, p= 0.03. There was no difference in RFS and OS by ADI. There was a trend towards decreased TNT (9 vs. 11 months, p=0.4) among BC pts living in areas with higher vs. lower ADI (>60% vs <60%). Notably, among stage 4 pts, in disadvantaged areas with ADI >60%, there was higher prevalence of Estrogen Receptor1(ESR1) 72% (8/11) vs. 27% (3/11) and TP53 75% (9/12) vs. 25% (3/12) somatic mutations compared to less disadvantaged areas with ADI <60%, p= 0.08 and 0.04, respectively. Conclusions: Our study shows disparities in access to germline testing, with lower rates among pts from more deprived areas. Moreover, somatic testing results show distinct mutational profiles across different ADI groups, suggesting potential impact of ADI on somatic characteristics. Larger studies are needed to investigate the association of ADI with somatic mutation burden of BC.

Predictors of endocrine immune-related adverse events (irAEs) in patients treated with immune checkpoint inhibitors (ICI) in routine clinical practice.

e14713 Background: Endocrine irAEs have been reported in association with various ICI and are characterized by their correlation with improved survival outcomes. A growing clinical comprehension of endocrine irAEs has facilitated the development of effective treatment strategies involving hormone replacement, which has been shown to enhance patients' quality of life. This study aims to evaluate outcomes in patients (pts) experiencing endocrine irAEs and identify predictive factors for their occurrence. Methods: This retrospective cohort study comprised 214 adult pts with solid malignant tumors who underwent ICI, either as monotherapy or in combination with other chemotherapeutic agents, at the SBIH Moscow Clinical Scientific and Practical Center named after A.S. Loginov of DHM. Results: Between June 2016 and January 2023, 214 pts were included in the study (46.3% female, 53.7% male). The median age was 62.6 years (range 32–91). At the data cut-off (1 February 2024) the median overall survival (OS) for the analyzed population was 34,5 months (95%CI 21,15-43,91). No significant difference in median OS was observed between pts with overall endocrine irAEs and those exhibiting specific manifestations, such as hypocortisolism and destructive diabetes mellitus. However, thyroid irAEs were associated with improved median OS: 27.5 months compared to 22.6 months in those without (p=0.037); RR=1.39 (95%CI 1.025-1.89; p=0.036). The grade of endocrine irAEs did not impact the survival benefit. A significant association was observed between age group and endocrine irAEs (Pearson’s χ2=9.92, p=0.019), with a left-skewed age distribution due to a higher proportion of participants aged 32–44 years among those developing endocrine irAEs. Pts with nodular goiter had a 2.91 times higher risk of developing thyroiditis (95%CI 1.03–8.25), although this did not reach statistical significance (p=0.067). Calcium channel blockers were associated with an increased risk of developing thyroid irAEs (RR=1.54, 95%CI 1.09–2.19; p=0.036). Antihistamine use was a robust predictor of hypothyroidism with a RR=1.68 (95%CI 1.05–2.7, p=0.04). Initial derived neutrophil-to-lymphocyte ratio (dNLR) ≥3 predict the development of endocrine irAEs (Pearson χ2=4.02; p=0.045) as well as dNLR>2.2 (Pearson χ2=4.22, p=0.04). Conclusions: Thyroid irAEs are correlated with improved OS. Predictors of endocrine irAEs may include young age (32-44 years), calcium channel blocker or antihistamine use, dNLR ≥3, or dNLR >2.2. Prospective trials are warranted to validate these findings.

Retrospective study of BRAF inhibitor-based therapy in patients with BRAF mutated advanced solid tumors (RBAST study).

e23265 Background: BRAF mutations are frequently found in melanoma, non-small cell lung cancer (NSCLC), thyroid cancer, to a lesser degree in other tumor types. These mutations constantly activate the kinase domain and result in MAPK pathway hyperactivation. Successful inhibition of this pathway with BRAF/MEK inhibitors (BRAFi/MEKi) results in a clinically meaningful benefit for some patients. The RBAST study, to assess the efficacy and survival outcomes of BRAF inhibitor-based therapy were investigated in patients with solid tumors harboring BRAF mutations. Methods: Data was collected on a total of 697 cases of solid tumor throughout China between Sep 2018 and Dec 2023. Next Generation Sequencing (NGS) is based on the sequencing of tissue samples. Among the samples, the primary endpoints of this study were objective response rate (ORR), disease control rate (DCR), progress-free survival (PFS) and overall survival (OS). The impact of co-mutations with other genes and BRAF variant allele frequency on treatment efficacy was also assessed in this study. Results: BRAF mutations were detected in 36 cases and 15 patients received BRAFi based therapy were identified. For patients who received BRAFi therapy, the best ORR was 26.7% with 4 patients recorded partial response (PR). There were 2 patients with progressive disease giving a DCR of 86.7%. Responses were observed in 4 types of tumors. PR was observed in patients with colorectal cancer (CRC) and melanoma. Median PFS (mPFS) and median OS (mOS) of these patients were 6.0 months (95% CI: 2.3-7.0) and 17.1 months (95% CI: 4.4-NA), respectively. The co-occurring mutation rates of TP53, PTEN, FAT2 and NF1 were 40% (6/15), 20% (3/15), 20% (3/15) and 13.3% (2/15), respectively. No significant difference (p>0.05) in PFS and OS was observed in any co-occurrence of mutations. BRAF mutation variant allele frequency is not an independent prognostic factor for survival with patients received BRAF inhibitor treatment (p>0.05). Conclusions: To the best of our knowledge, this is the first retrospective study exploring BRAFi treatment effectiveness to BRAF mutated solid tumors for Chinese patients. Treatment with BRAF inhibitors is effective for BRAF mutated advanced solid tumors with DCR of 86.7%, mPFS of 6.0 months and mOS of 17.1 months. Co-mutations with other genes and BRAF variant allele frequency did not impact survival outcomes.

Use of Gynecologic Cancer Intergroup (GCIG) CA125 criteria to evaluate cell cycle checkpoint kinase 1 inhibitor (CHK1i) prexasertib response and disease progression (PD) in recurrent high-grade serous ovarian cancer (HGSOC).

e17536 Background: Data suggest discordance between serum biomarker CA125 and Response Evaluation Criteria in Solid Tumors (RECIST) response and/or PD in chemotherapies and PARP inhibitors. Here, we investigated the impact of CA125 on prognosis and its relationship with RECIST response/PD in HGSOC patients treated with a CHK1i, prexasertib. Methods: All relapsed HGSOC patients in this post-hoc analysis were treated with prexasertib at 105 mg/m2 IV every 14 days in a 28-day cycle until RECIST v1.1 PD, unacceptable toxicity, or consent withdrawal (NCT02203513). CA125 was assessed every cycle and CT scans obtained every two cycles. GCIG CA125 response was defined as >50% reduction confirmed after 28 days and PD as > 2X upper limit of normal (ULN) if baseline was normal or doubling of nadir if baseline was elevated. Patients with baseline CA125 < 2X ULN or nadir > 750 units/mL were not eligible for analysis. Clinical benefit rate (CBR; CR + PR + SD ≥ 6 months) was analyzed by Fisher’s exact test. Progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier curves and univariate (UV) and multivariate (MV) hazard ratios by Cox regression models. Positive predictive value (PPV) and negative predictive value (NPV) confidence intervals (CI) were estimated by the Wilson-Brown method. Results: Eighty-one patients received at least two cycles of prexasertib between January 2015 and November 2020, out of which 72 were evaluable by both RECIST and GCIG CA125 response criteria. CBR in the CA125 response (CA125-R) group (24/32; 75%) was significantly higher than the CA125 non-responder (CA125-N) group (12/40; 30%; p<0.001). CA125 response tended to occur prior to RECIST response (median: 28 days; IQR: 28, 30). PFS was improved in CA125-R at 8.0 months vs. CA125-N at 3.5 months (UV HR: 0.34, 95% CI: 0.21-0.56; MV HR: 0.35, 95% CI: 0.20-0.59). OS was improved in CA125-R at 19.8 months vs. CA125-N at 10.3 months (HR: 0.43, 95% CI: 0.26-0.71; MV HR: 0.45, 95% CI: 0.27-0.75). No significant differences in PFS or OS were observed between subgroups by BRCA mutation status, platinum-sensitivity, FIGO stage, baseline CA125 value, or previous PARP inhibitor therapy in MV analysis. Out of 70 patients evaluable by RECIST and CA125 PD criteria, 43 patients had CA125 PD and 24 of those had RECIST PD occur within one cycle of CA125 PD (PPV: 59%; 95% CI: 43-72). Of the 27 patients without a CA125 PD, 9 patients also did not have a RECIST PD (NPV: 33%; 95% CI: 19-52). Conclusions: GCIG CA125 response is an independent predictor of response to CHK1i in HGSOC prior to the first restaging scans and is associated with a higher CBR and improved PFS and OS. However, GCIG CA125 PD has poor PPV and NPV within a one cycle window, indicating that CA125 alone is not sufficient in monitoring for PD in HGSOC treated with CHK1i.

Real-world treatment outcomes in patients with HR+ HER2- advanced breast cancer treated with CDK4/6 inhibitors and endocrine therapy.

1067 Background: The addition of cyclin-dependent kinases 4 and 6 inhibitors (CDK4/6i) to endocrine therapy (ET) improves progression-free (PFS) and overall survival (OS) in hormone receptor-positive, HER2-negative advanced breast cancer (HR+/HER2- ABC) in 1st and 2nd line setting. The international guidelines advise using 1st line CDK4/6i; however, its superiority over 2nd line use is debatable. Methods: We retrospectively analyzed clinical data of HR+/HER2- ABC pts administered palbociclib (PAL), ribociclib (RIB) or abemaciclib (ABM) added to either 1st or 2nd line ET. Results: A total of 1511 pts (PAL=507; RIB=724; ABM=280) from 11 cancer Polish centers treated between Sep 2017 and Dec 2023. Median follow-up was 29.6 (95%CI 28.2 to 30.9) months (m). Mean age was 62.6±11.9, 58.9±12.2 and 61.4±10.8 years in PAL, RIB and ABM, respectively. 815 patients (53.9%) had visceral and 1109 (73.4%) bone metastases. 1254 pts (83%) received 1st line CDK4/6i, in 841 (67%) combined with aromatase inhibitor (AI) and in 413 (33%) with fulvestrant (FUL). Second-line CDK4/6i was used in 257 pts (17%); 59 (23%) and 198 (67%) with AI and FUL, respectively. There was no difference in PFS, PFS2, and OS between three CDK4/6i in combination with AIs in 1st and 2nd line setting. Median PFS (95%CI) for 1st line AI with PAL, RIB, ABM were 29.0 (21.6-37.0), 32.0 (24.5-36.9), not reached (N/R) (21.9-N/R) m; PFS2: 37.0 (28.8-N/R), 33.0 (29.5-40.2), N/R (28.9-N/R); OS: 48.9 (42.9-N/R), 48.6 (41.9-N/R), N/R (36.7-N/R). Median PFS (95%CI) for 2nd line AI with PAL, RIB, ABM were 11.9 (9.49-N/R), 10.8 (8.27-N/R), 11.0 (6.43-N/R); PFS2: 22.7 (16.8-N/R), 32.3 (24.9-N/R), 31.0 (13.2-N/R); OS: 27.4 (24.4-N/R), 36.7 (30.1-N/R), 45.5 (29.0-N/R) (no statistical significance). In contrast, RIB+FUL was superior to PAL+FUL and ABM+FUL in both lines (Table). The safety profile was concordant with the reported data for ET + CDK4/6i. Conclusions: Identifying predictive factors of response is essential for the rational use of CDK4/6i and ET in HR+/HER2- ABC. [Table: see text]

Tumor treating fields for newly diagnosed high-grade glioma based on the criteria of 2021 WHO CNS5: A retrospective analysis of Chinese patients in a single center.

High-grade glioma (HGG) is known to be characterized by a high degree of malignancy and a worse prognosis. The classical treatment is safe resection supplemented by radiotherapy and chemotherapy. Tumor treating fields (TTFields), an emerging physiotherapeutic modality that targets malignant solid tumors using medium-frequency, low-intensity, alternating electric fields to interfere with cell division, have been used for the treatment of new diagnosis of glioblastoma, however, their administration in HGG requires further clinical evidence. The efficacy and safety of TTFields in Chinese patients with HGG were retrospectively evaluated by us in a single center. We enrolled and analyzed 52 patients with newly diagnosed HGG undergoing surgery and standard chemoradiotherapy regimens from December 2019 to June 2022, and followed them until June 2023. Based on whether they used TTFields, they were divided into a TTFields group and a non-TTFields group. Progression-free survival (PFS) and overall survival (OS) were compared between the two groups. There were 26 cases in the TTFields group and 26 cases in the non-TTFields group. In the TTFields group, the median PFS was 14.2 months (95% CI: 9.50-18.90), the median OS was 19.7 months (95% CI: 14.95-24.25) , the median interval from surgery to the start of treatment with TTFields was 2.47 months (95% CI: 1.47-4.13), and the median duration of treatment with TTFields was 10.6 months (95% CI: 9.57-11.63). 15 (57.69%) patients experienced an adverse event and no serious adverse event was reported. In the non-TTFields group, the median PFS was 9.57 months (95% CI: 6.23-12.91) and the median OS was 16.07 months (95% CI: 12.90-19.24). There was a statistically significant difference in PFS (p = 0.005) and OS (p = 0.007) between the two groups. In this retrospective analysis, TTFields were observed to improve newly diagnosed HGG patients' median PFS and OS. Compliance was much higher than reported in clinical trials and safety remained good.

Effect of indoximod-based chemo-immunotherapy in patients with pediatric brain tumors on activation and clonal proliferation of a circulating population of early non-exhausted stem-like CD8+ T cells whose on-treatment expansion is predictive of long-term outcome.

2566 Background: Combination chemo-immunotherapy is a promising strategy, but it is difficult to determine whether clinical responses are associated with on-target immune activation or just due to chemotherapy. The field lacks well-accepted readouts of on-treatment T cell activation/expansion to answer this question. Methods: We analyzed longitudinal blood samples from 30 patients with pediatric brain tumors treated with the IDO-inhibitor drug indoximod combined with either chemotherapy (n=27, NCT02502708, NCT04049669) or chemotherapy plus ibrutinib (n=3, NCT05106296). Patients in this “Training Set” were selected to include multiple trials, various histologies/molecular risk factors, and a range of outcomes (overall survival (OS) range 6-55 months). Longitudinal blood samples (2-11 samples/patient) were obtained over 4-36 months, depending on duration of treatment, and analyzed by single-cell RNA and TCR sequencing (scRNAseq/TCRseq). TCR clonotypes of interest were defined as having at least 2-fold expansion (or appearance de novo) during treatment compared to baseline. To calculate a “Clonal Expansion Index” (CEI), the total number of T cells belonging to treatment-expanded clonotypes was summed for each sample and expressed as a percentage of total T cells. The peak CEI value for each patient was used to stratify subjects into “Immune Responders” vs “Non-responders”, based on an optimized cutoff established by Receiver Operating Characteristic (ROC) analysis with Youden’s J statistic. Results: In these patients, CEI ranged from <1% to >60% of all circulating T cells. The optimized cut-point was found to be a CEI of 8.6%, producing a sensitivity of 91% and specificity of 77% for this dataset. Kaplan-Meier analysis showed a highly significant 3-fold difference in median OS for the CEI-High patients (26.5 months) compared to CEI-Low (8.9 months, p=0.0003). The CEI metric was far superior as a predictor of long-term outcome than radiographic response by RANO, RAPNO or iRANO criteria, which were not predictive. UMAP clustering and Monocle trajectory analysis of the treatment-expanded T cells revealed that the majority arose from a population of early stem-like cells (TCF7+ LEF1+ FOXP1+), progressing through a more activated stem-like population (HOBIT/ZNF683+) to attain proliferation and effector maturation states. Throughout this sequence, the T cells showed minimal markers of exhaustion (PDCD1, HAVCR2, TOX) and acquired a lytic effector phenotype. Conclusions: We hypothesize that clonal expansion of this population of non-exhausted stem-like T cells, as quantitated by the CEI assay, provides a mechanistically based pharmacodynamic readout of T cell response to indoximod-based (and perhaps other) chemo-immunotherapy.

Prognosis of deep margin excisions within or beyond subcutaneous fat for invasive acral melanoma of the sole: A multi-institutional retrospective study.

9582 Background: There is scant evidence concerning the optimal deep margins for wide excision (WE) of primary tumors in melanoma surgery. Despite the importance of preserving plantar subcutaneous fat in the sole for cushioning function, no study data exists on the optimal depth margins for acral melanoma of the sole (sole AM). Thus, we aimed to compare the prognosis associated with different deep margin excisions for sole AM. Methods: We conducted a retrospective review of clinical records from patients with invasive resectable sole AM who underwent WE of the primary tumor across 41 Japanese institutions. The prognosis was compared between two groups based on different deep margin excisions: those with tumors excised within (S-group) or beyond (D-group) the subcutaneous fat. Kaplan—Meier analysis and multivariable Cox proportional hazard models were used to estimate the survival probabilities. Propensity-score matching (PSM) was used to adjust for numerical differences in baseline characteristics. Results: This study included a total of 425 patients (median age 73 years) with invasive sole AM, who underwent WE based on the peripheral surgical margins recommended in the National Comprehensive Cancer Network guidelines. The median follow-up period was 48 months. The S-group included 263 patients and the D-group 162 patients. The baseline characteristics significantly differed between the two groups in terms of Breslow thickness ( P < 0.01), presence of ulcer ( P < 0.01), and presence of in-transit, satellite, and regional lymph node metastasis ( P < 0.01). The other baseline characteristics were similar. Although local recurrence-free survival (LRFS) did not significantly differ between groups, the S-group showed significantly longer disease-free (DFS) and overall survival (OS) than the D-group (5-year LRFS, 96% vs. 90%; P = 0.08; 5-year DFS, 70% vs. 41%; P < 0.01; 5-year OS, 84% vs. 69%; P < 0.01). In contrast, multivariable Cox proportional hazard models showed no significant differences in LRFS, DFS, and OS between the two groups (LRFS, hazard ratio [HR], 1.87, P = 0.18; DFS, HR, 1.05, P = 0.77; OS, HR, 1.08, P = 0.72). After PSM, the groups were balanced at a 1:1 ratio (128 patients in each group) and no statistical significance was found in LRFS, DFS, or OS between the adjusted groups (5-year LRFS, 97 % vs. 91 %; P = 0.40; 5-year DFS, 65 % vs. 48 %; P = 0.053; 5-year OS, 80 % vs. 73%; P = 0.18). Subgroup analysis of stage I, II, and III also indicated no statistical significance in LRFS, DFS, and OS between the two groups (stage I: LRFS, P = 0.11; DFS, P = 0.65; OS, P = 0.26; stage II: LRFS, P = 0.37; DFS, P = 0.06; OS, P = 0.85; stage III: LRFS, P = 0.29; DFS, P = 0.67; OS, P = 0.85). Conclusions: In sole AM, WE with excessive deep margins beyond the subcutaneous fat did not improve prognosis. Excision of the primary tumor within the subcutaneous fat may be an optimal deep margin for sole AM.

Efficacy of immunotherapy retreatment in patients with non-small cell lung cancer receiving consolidative durvalumab.

8625 Background: The standard of care for patients with stage III, unresectable non-small cell lung cancer (NSCLC) is curative intent chemoradiation followed by durvalumab consolidation. However, most patients progress necessitating further treatment. Although immunotherapy (IO) is the cornerstone of treatment in advanced/metastatic disease, the benefit in patients who received prior durvalumab is unclear. Clinical trials excluded prior IO exposure and current guidelines do not provide specific recommendations. This study aims to characterize the efficacy of post-durvalumab IO treatment. Methods: This was a retrospective cohort study conducted at the University of North Carolina (UNC) Medical Center. Adult patients with unresectable NSCLC who received chemoradiation and consolidative durvalumab followed by subsequent systemic therapy for disease progression between February 2018 to May 2023 were included. Patients were excluded if they received targeted therapy as subsequent systemic treatment. The primary endpoint was real-world progression-free survival (rwPFS) of post-durvalumab treatment. Key secondary outcomes included physician assessed response, overall survival (OS), and subgroup analyses to identify biomarkers which may predict benefit from IO retreatment. Results: A total of 66 patients met inclusion criteria, of which 39 (59%) received subsequent chemotherapy alone and 27 (41%) received IO-containing therapy. Of those who received IO-containing treatments, 18 received chemotherapy plus IO (chemoIO) and 9 received IO alone. Durvalumab was discontinued for therapy completion (chemotherapy: 18%, IO: 26%), disease progression (chemotherapy: 54%, IO: 48%), or side effects (chemotherapy: 26%, IO: 19%). No significant difference was observed between median rwPFS of chemotherapy versus IO-containing therapy (7.0 months vs 5.8 months, respectively; p = 0.9419) or OS (13.0 months vs 12.6 months, respectively; p = 0.5769). Notably, patients in our cohort who received IO alone had a numerically longer rwPFS compared to chemoIO (12.6 months vs 5.8 months, respectively; p = 0.3951), as well as a numerically longer OS (12.6 months vs 6.4 months, respectively; p = 0.8885). Conclusions: This study found that there were no statistically significant differences in median rwPFS or OS between chemotherapy or IO-based regimens following the conclusion of durvalumab consolidation. Interestingly, we observed a numerically improved rwPFS and OS in those who received IO alone compared to chemoIO, which suggests there may be key clinical or molecular features that may help identify patients who benefit from IO therapy. We acknowledge the limitations of this small, retrospective study. The efficacy of IO-based regimens post-durvalumab warrants further investigation in a larger, prospective manner to assess the risks and benefits of IO retreatment.

A retrospective study of PD-1/PD-L1 combination therapy versus TKI monotherapy in elderly patients with uHCC.

e16191 Background: The rising elderly population worldwide has led to an increase in cases of liver cancer among individuals aged ≥65 years. With the advancement of medical treatments, the management of unresectable hepatocellular carcinoma (uHCC) now predominantly involves immunotherapy combination therapy. However, existing data in elderly patients are limited. Therefore, we conducted a retrospective analysis on the efficacy and safety of immune combination therapy compared with TKI monotherapy in elderly uHCC patients. Methods: A retrospective analysis was conducted on uHCC patients aged ≥65 years at the First Hospital of Jilin University between March 2019 and July 2023. The targeted therapy used in this study included, but not limited to, sorafenib, lenvatinib. PD-1/PD-L1 inhibitors such as pembrolizumab and atezolizumab were also utilized. Follow-up was conducted until death, loss to follow-up, or December 2023. The primary endpoint was overall survival (OS), while the secondary endpoints were progression-free survival (PFS), objective response rate (ORR), disease control rate (DCR), adverse events (AEs) and immune-related AEs (irAEs). Results: A total of 70 patients were included in this study, with 33 patients receiving monotherapy, and 37 patients receiving immune combination therapy. The median age was 68 years old, with 78.8% being male. The combination therapy group had a higher ORR (10.8% vs. 3.0%) DCR (48.5% vs. 45.9%) compared to the monotherapy group. However, there were no statistically significant differences in median PFS (7.0 vs. 12.0 months) and median OS (9.8 vs. 17.5 months) between the two groups. Among the subgroups, combination therapy showed better results except for patients with ascites. Immune combination therapy appeared to be superior to monotherapy in terms of both PFS (6.2 vs. 6.8 months, P = 0.888) and OS (6.8 vs. 10.2 months, P = 0.888). Multivariate analysis showed that elevated total bilirubin may be a risk factor for OS in patients ≥65 years old. Regarding safety, the adverse event (AE) rate was 39.3% in the monotherapy group. The most common AEs reported were abdominal distension (18.2%), fatigue (15.2%), and proteinuria (6.1%). In the combination therapy group, the AE rate was higher at 59.4%, with the most common AEs being thyroid function abnormalities (24.3%), fatigue(24.3%), decreased appetite(18.9%), and hypertension(18.9%). The rate of drug withdrawal due to adverse reactions was significantly higher in patients receiving combination therapy compared to those receiving monotherapy (40.5% vs. 12.1%, P = 0.008). Conclusions: In elderly uHCC patients (≥65 years old), immune combination therapy shows a higher ORR and potentially longer PFS and OS except patients with ascites. AEs and drug withdrawal rates are higher in the combination therapy group. Further discussions are needed to determine the optimal treatment strategy for elderly uHCC patients.

Real-world experience with CDK4/6 inhibitors in the first-line palliative setting for HR+/HER2- advanced breast cancer.

e13066 Background: CDK4/6 inhibitors paired with aromatase inhibitors (AIs) have redefined the first-line treatment for metastatic breast cancer, as evidenced by several pivotal trials. Although progression-free survival (PFS) has been consistent across these studies, notable differences in overall survival (OS) have been observed. Our study evaluates the real-world efficacy of palbociclib and ribociclib, in combination with AIs, for hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) advanced breast cancer at our institution. Methods: We retrospectively reviewed charts of adult patients with advanced metastatic breast cancer at a single academic center from January 1, 2015, to December 1, 2022. We extracted relevant clinical data to estimate the median PFS and OS associated with palbociclib and ribociclib. Results: Among 75 patients analyzed, the majority were female (98.7%) and postmenopausal (77.3%). Palbociclib was prescribed to 74.7%, while 52.0% presented with de novo stage IV disease. Those on ribociclib were significantly younger (median age 57.6 vs. 67.5 years; p = 0.013) and more likely premenopausal (42.1% vs. 5.4%; p < 0.001). Median PFS for palbociclib was 20.3 months (95% CI: 14.8-46.0), with an OS of 37.2 months (20.3 months to not reached). Median PFS and OS for ribociclib were not reached. The differences in PFS and OS between the drugs were not statistically significant (PFS log-rank p = 0.1; OS log-rank p = 0.2). Conclusions: Our real-world analysis suggests a non-significant trend towards improved PFS and OS for ribociclib over palbociclib when used with AIs for treating HR+/HER2- advanced breast cancer, potentially influenced by the younger age of patients on ribociclib. Further research with larger cohorts and extended follow-up is warranted to substantiate these observations.

Start low, go slow: A systematic review and meta-analysis of upfront tailored treatment dosing in older adults with advanced cancer.

1625 Background: Older adults aged ≥65 with advanced cancer are underrepresented in treatment trials. This gap limits informed decision-making regarding the dose and tolerability of proposed therapies. Standard chemotherapy is associated with >50% serious adverse event rate among older adults, with frequent dose reduction and early treatment discontinuation. A "start low, go slow" (SLGS) approach begins systemic therapy at lower-than-standard doses and increases the dose if well-tolerated. This alternative dosing strategy has shown value in minimizing adverse events and functional decline without compromising the overall effectiveness of treatments. We conducted the first systematic review and meta-analysis of SLGS effectiveness among older adults across advanced cancers. Methods: This review, was registered with PROSPERO and adhered to PRISMA criteria, covering PubMed, Journal of Geriatric Oncology and EMBASE from January 2000 until December 15, 2024. Eligible study designs were randomized controlled trials, retrospective trials, and non-randomized clinical trials with patients with advanced cancers who underwent systemic therapy. We reported all studies that studied the SLGS approach. Our studied outcomes were overall survival (OS), progression-free survival (PFS), treatment discontinuation, and toxicity. Data extraction included author, patient, cancer type, treatment, and survival outcomes. A meta-analysis using fixed effects and risk ratios (RR) was performed when sufficient data was available; significant outcomes had α<0.05. Results: Our search identified 13 studies testing SLGS strategies in oncology, including 3,508 patients, with a median age 63-78 years. The -represented cancers included colorectal (6 studies, 3059 patients total), lung (2 studies, 113 pts), chronic myeloid leukemia (2 studies, 127 pts), non-Hodgkin lymphoma (1 study, 45 pts), and prostate (2 studies, 164 pts) cancers. Ten (77%) studies assessed OS and PFS. Five studies (39%) compared SLGS against standard doses, finding no significant differences in PFS and OS across all trials. Dose escalation rates for SLGS ranged from 5% to 60%. The ability to complete planned cycles was higher with SLGS compared to standard dose (1 study, 43% vs 26%, p=0.04). Treatment discontinuation was not different for SLGL vs. standard dose (5 studies, meta-analysis RR 1.07, 95% CI 0.90-1.27, p=0.42). Toxicity ranged from 5% to 89% across studies; SLGS had lower grade ≥3 adverse events compared to standard dose (4 studies, meta-analysis RR 0.88, 95% CI 0.80-0.94, p < 0.001). Conclusions: This is the first systematic review and meta-analysis analyzing the SLGS approach to systemic therapy dosing in older adults with advanced cancer. Compared to standard-dose systemic therapy, older adults pursuing a SLGS strategy had greater completion of planned cycles, reduced toxicity, and similar survival.

SU2C-SARC032: A randomized trial of neoadjuvant RT and surgery with or without pembrolizumab for soft tissue sarcoma.

11504 Background: Surgery & radiation therapy (RT) yield high local control rates for soft tissue sarcoma (STS) of the extremity and limb girdle. However, patients (pts) with high-grade stage III STS are at significant risk for developing metastasis. Median survival for pts with metastatic STS is < 2 years. SARC028 (NCT02301039) evaluated the efficacy of pembrolizumab (pembro) for metastatic STS, showing 20% and 8.7% response rates in undifferentiated pleomorphic sarcoma (UPS) and pleomorphic/dedifferentiated liposarcoma (LPS), respectively. We hypothesized that neoadjuvant pembro with concurrent RT followed by surgery and adjuvant pembro for stage III UPS, including myxofibrosarcoma, or LPS would stimulate an anti-tumor immune response to eliminate micrometastatic disease & improve disease-free survival (DFS). SU2C-SARC032 (NCT03092323) is a multi-institutional, international, randomized phase 2 trial evaluating the safety and efficacy of adding pembro to standard of care (SOC) RT & surgery for pts with stage III UPS or LPS. Methods: Pts aged > 12 yo with stage III (FNCLCC grade 2 or 3) UPS or LPS of the extremity and limb girdle were enrolled. Pts were randomized (1:1, stratified by grade) to neoadjuvant RT (50 Gy/25 fx) then surgery (SOC arm) or neoadjuvant pembro and RT then surgery & adjuvant pembro (EXP arm). Pembro was given 200 mg IV Q3 wk for 3 doses (before, during & after RT) & up to 14 adjuvant cycles. The primary endpoint was 2-yr DFS. Secondary endpoints included local recurrence-free survival (LRFS), distant disease-free survival (DDFS), & overall survival (OS). Target enrollment of 126 evaluable patients (max 144 total) provided 80% power (1-sided α = 0.05) to distinguish between a null hypothesis of 50% 2-yr DFS rate & alternative hypothesis of 75% 2-yr DFS rate by log-rank test, with initial analysis at 45 DFS events. Cox models were stratified by grade; primary analysis was a one-sided stratified log-rank test. Results: Between July 2017-November 2023, 143 patients were enrolled, predominantly with UPS (85%) & grade 3 (64%) histology. Median follow-up for alive patients is 24.1 mo. DFS in the EXP arm is significantly higher than the SOC arm (p = 0.023; HR 0.57, 90% CI: 0.35, 0.91). Estimated 2-yr DFS is 53% (90% CI: 43, 66%) for SOC vs 70% (90% CI: 61, 81%) for EXP arm. Currently, there is no statistically significant difference in LRFS (HR 0.55, 95% CI: 0.21, 1.42), DDFS (HR 0.57, 95% CI: 0.32, 1.01), or OS (HR 0.39, 95% CI: 0.14, 1.12). Pts with grade 3 sarcomas had improved DFS with pembro (HR 0.47, 95% CI: 0.25, 0.89), but no difference in DFS was observed in grade 2 tumors (HR 1.21, 95% CI: 0.35, 4.18). The proportion of patients with grade 3+ adverse events was significantly higher in EXP (52%) vs SOC arm (26%) (p = 0.002). Conclusions: Theaddition of neoadjuvant & adjuvant pembro to RT and surgery significantly improves DFS for pts with stage III UPS and LPS of the extremity and limb girdle. Clinical trial information: NCT03092323 .

Long-term follow up of E3311, a phase II trial of transoral surgery (TOS) followed by pathology-based adjuvant treatment in HPV-associated (HPV+) oropharynx cancer (OPC): A trial of the ECOG-ACRIN Cancer Research Group.

6009 Background: E3311 is a phase II trial of TOS by credentialed surgeons with pathology-driven deintensified post-operative management in HPV+ OPC. Intermediate risk patients were randomized between standard and reduced dose radiation. We present mature outcome data, at median follow up of 52.4 months (m). Methods: Patients were eligible who had resectable cT1-2 stage III/IV AJCC7 p16+ OPC without matted neck nodes. Those with clear margins, 0-1 + nodes (LN), and no extranodal extension (ENE) were observed (Arm A, N=38); those with clear margins, 2-4 + LN, or ENE ≤1mm were randomized to 50Gy (Arm B, N=100) or 60Gy (Arm C, N=108); those with involved margins, >4 + LN, or >1mm ENE received weekly cisplatin 40 mg/m2 and 60-66Gy (Arm D, N=113). Progression-free survival (PFS) and overall survival (OS) were estimated by the Kaplan-Meier method and compared using a log-rank test, stratified by arm for comparisons of primary site and smoking history. Results: Among 359 evaluable patients, 54-m PFS and OS were 90.6% (90% CI: 87.2%, 93.1%) and 95.3% (90% CI: 93.0%, 96.9%). 54-m PFS by arm was: A 93.2% (90% CI: 79.6%, 97.8%); B 94.9% (90% CI: 89.7%, 97.5%); C 90.2% (90% CI: 82.7%, 94.6%) and D 85.5% (90% CI: 77.5%, 90.8%). 54-m OS by arm was: A 97.1% (90% CI: 85.7%, 99.4%); B 97.9% (90% CI: 93.5%, 99.3%), C 95.1% (90% CI: 90.1%, 97.6%) and D 92.5% (90% CI: 86.9%, 95.7%). Among patients in Arm A, 11 had N0 and 27 N1 stage. Median Arm A lymph node yield (LNY) was 29 (11 to 91 LN) and did not differ for patients with/without recurrence (p=0.83). All Arm A recurrences were in N1 patients: 1 at 18 m and 3 at > 40 m. No significant difference in PFS or OS was observed by prescribed radiation dose for intermediate risk patients (Arm B vs. C). Outcome did not differ by primary site of tonsil vs. other OPC (oOPC): 54-m PFS Tonsil 89.3% (90% CI: 84.9%, 92.5%) vs. oOPC 92.9% (90% CI: 87.0%, 96.2%), p=0.28, and 54-m OS Tonsil 94.6% (90% CI: 91.5%, 96.6%) vs oOPC 96.6% (90% CI: 92.4%, 98.5%), p=0.35. Smoking history also did not impact outcome. 54-m PFS was 89.9% (90% CI: 85.9%, 92.9%) for those with ≤10 pack-years (PY) tobacco exposure and 91.7% (90% CI: 83.9%, 95.9%) for those with >10 PY. Conclusions: TOS, neck dissection with deintensified risk-based post-operative management results in outstanding 54-m PFS and OS across all subsites of T1-2 p16+ OPC, irrespective of smoking history. Late recurrence is not increased when post-operative radiation is reduced from 60 to 50Gy for intermediate risk patients. Pathologic >1mm ENE (N=87), involved margin (N=12) or >4 involved LN (N=30) accurately identify patients at increased recurrence risk and outcomes were favorable among such patients. Among patients with favorable pathologic characteristics, a subset with N1 disease are at risk for late recurrence and further characterization of these patients is warranted. Clinical trial information: NCT01898494 .

Comparing the efficacy and safety of TACE combined with donafenib or lenvatinib in the treatment of unresectable hepatocellular carcinoma.

e16153 Background: For unresectable Intermediate/Advanced-Stage Hepatocellular Carcinoma (uHCC), the efficacy of TACE monotherapy has been unsatisfactory. Studies have shown that TACE combined with molecular targeted drugs exhibits potential synergistic effects, which can prolong the overall survival (OS) of uHCC patients (pts). Donafenib and lenvatinib are small molecule targeted drugs approved as first-line treatments for HCC. The aim of this study is to evaluate the efficacy and safety of TACE combined with donafenib or lenvatinib in the treatment of uHCC. Methods: A retrospective analysis was conducted on the clinical data of uHCC pts treated with donafenib combined with TACE (Dona-TACE) or lenvatinib combined with TACE (Lenva-TACE) at The First Affiliated Hospital of USTC from January 2018 to October 2022. The OS, progression free survival (PFS), objective response rate (ORR), and disease control rate (DCR) of two groups of pts were evaluated according to the mRECIST, and the incidence of adverse events (AEs) was evaluated using CTCAE 5.0. Pts with complete baseline information and at least one follow-up evaluation were incorporated in the final analysis. Results: A total of 105 pts treated with donafenib and 59 pts treated with lenvatinib met the study criteria. After matching with a 1:1 propensity score, 35 pts in the Dona-TACE group and 35 pts in the Lenva-TACE group were included in this study. The baseline characteristics of the two groups are similar (males, 85.7% vs. 85.7%; ECOG PS of 1, 74.3% vs. 71.4%; Exceeding up-to-seven criteria, 74.3% vs. 62.9%; BCLC stage C, 42.9% vs. 40.0%; Combination immunotherapy, 62.9% vs. 54.3%). There was no significant difference in median OS (29.0 vs. 27.3m, P = 0.748) and median PFS (10.5 vs. 10.3m, P = 0.869) in the Dona-TACE and Lenva-TACE groups. The ORR (85.7% vs. 74.3%, P = 0.969) and DCR (94.3% vs. 97.1%, P = 0.280) of the Dona-TACE and Lenva-TACE group were equivalent. Compared with the Lenva-TACE group, pts in the Dona-TACE group had a lower incidence of grade 3-4 adverse events (any grade, 94.3% vs. 91.4%; grade 3 or 4, 34.3% vs. 57.1%, P < 0.05). Among them, pts in the Dona-TACE group had a lower incidence of proteinuria (any grade, 22.9% vs 42.9%; grade 3 or 4, 0 vs17.1%, P < 0.05) and hypertension (any grade, 42.9% vs 71.4%; grade 3 or 4, 2.9% vs 25.7%, P < 0.05) than those in the Lenva-TACE group. Multivariate analysis showed that ECOG PS score, up-to-seven criteria, and tumor blood supply were independent prognostic factors affecting OS (HR, 3.470, 3.619, 3.067; P < 0.05). Conclusions: The efficacy of donafenib and lenvatinib combined with TACE in the treatment of uHCC is equivalent. Furthermore, compared to lenvatinib combined with TACE, donafenib combined with TACE has higher safety.

Clinicopathological features and prognosis for pregnancy-associated breast cancer: A matched case-control study.

e12573 Background: Pregnancy-Associated Breast Cancer (PABC) is defined as breast cancer diagnosed during pregnancy or within 1 year postpartum. Studies indicate that PABC tends to have poorer prognostic characteristics and often worse outcomes compared to non-PABC. The primary goal of this study is to examine disease characteristics associated with PABC versus non-PABC in the United Arab Emirates (UAE), where breast cancer is characterized by a younger age and more advanced stage compared to North America and Europe. Additionally, the UAE has a younger mean age for first-time mothers and a similar age for all mothers compared to North America and Europe. Methods: Patients with PABC were selected from BC cases of women ≤40 years treated at Tawam Hospital, the largest tertiary cancer center in the UAE between January 2003 – December 2020. Each PABC patient was matched to two non-PABC patients (controls) according to cancer stage, age, and year of diagnosis. The Fisher's exact and Kruskal-Wallis tests were used to compare both study groups. Results: 66 women with PABC were identified, with a median age at diagnosis of 33.0 (interquartile range [IQR]: 30.0-37.0) years. Most presented with stage II (45.5%) or stage III (39.4%) disease and underwent mastectomy and axillary lymph node dissection. Invasive ductal carcinoma was the most frequent histological type in both groups (86.4% for PABC and 85.6% for non-PABC). PABC was associated with higher grade 3 disease compared with non- PABC (53% vs 51.5% p= 0.009). More women with PABC received neoadjuvant chemotherapy (NACT) and radiation therapy (RT) compared with non-PABC (48.5% vs 3%, p <0.001 and 80.3% vs 65.9%, p= 0.046, respectively). There were no statistically significant differences with regards to tumor size, nodal involvement, underlying histologic subtype, ER/PR expression or HER-2 status, rates of breast-conserving surgery and adjuvant chemotherapy receipt between the two groups. The median progression-free survival (PFS) was 3.2 years (IQR: 1.9-7.1) in PABC and 3.4 years (IQR: 1.2-6.2) in non-PABC group (P=0.306). The median overall survival (OS) was 4.5 years in the PABC (IQR: 2.4-7.9) versus 4.5 years (IQR: 1.9-7.9) in the non-PABC group (P = 0.786). Conclusions: PABC in women ≤40 years appears to be associated with grade III tumors and NACT and RT receipt compared to non-PABC. However, there were no significant difference in PSF and OS between PABC and non-PABC. Further analysis is underway to evaluate the prognostic factors and their effect on survival.

Comparison of the Prognostic Value of C-Reactive Protein to Albumin Ratio and Glasgow Prognostic Score in Patients with Diffuse Large B-Cell Lymphoma

To compare the prognostic value of two predictive models based on C-reactive protein (CRP) and albumin (ALB), namely the CRP to ALB ratio (CAR) and the Glasgow prognostic score (GPS), in newly diagnosed patients with diffuse large B-cell lymphoma (DLBCL). The data of newly diagnosed DLBCL patients admitted to our center from May 2014 to January 2022 were reviewed. A total of 111 patients who completed at least 4 cycles of R-CHOP or R-CHOP-like chemotherapy with detailed clinical, laboratory data and follow-up information were included. The receiver operating characteristic (ROC) curve was performed to evaluate the predictive value of pre-treatment CAR on disease progression and survival. Furthermore, the association between CAR and baseline clinical, laboratory characteristics of patients was evaluated, and progression-free survival (PFS) and overall survival (OS) were compared between different CAR and GPS subgroups. Finally, the univariate and multivariate COX propor-tional hazard regression models were used to analyze the factors affecting disease outcomes. ROC curve showed that the area under the curve (AUC) of CAR predicting PFS and OS in DLBCL patients was 0.687 (P =0.002) and 0.695 (P =0.005), respectively, with the optimal cut-off value of 0.11 for both predicting PFS and OS. Compared with the lower CAR (<0.11) group, the higher CAR (≥0.11) group had more clinical risk factors, including age >60 years (P =0.025), ECOG score ≥2 (P =0.004), Lugano stage III-IV (P < 0.001), non-germinal center B-cell-like (non-GCB) subtype (P =0.035), elevated lactate dehydrogenase (LDH) ( P < 0.001), extranodal involved site >1 (P =0.004) and IPI score >2 (P < 0.001). The interim response evaluation of patients showed that the overall response rate (ORR) and complete response rate (CRR) in the lower CAR group were both significantly better than those in the higher CAR group (ORR: 96.9% vs 80.0%, P =0.035; CRR: 63.6% vs 32.5%, P =0.008). With a median follow-up of 24 months, patients with lower CAR had significantly longer median PFS and OS than those with higher CAR (median PFS: not reached vs 67 months, P =0.0026; median OS: not reached vs 67 months, P =0.002), while there was no statistical difference in PFS (P =0.11) and OS (P =0.11) in patients with GPS of 0, 1, and 2. Multivariate Cox regression analysis indicated that only sex (male) and IPI score >2 were independent risk factors for both PFS and OS. CAR is significantly correlated with disease progression and survival in DLBCL patients; And compared with GPS, CAR has more advantages in predicting disease outcomes in DLBCL patients.

Impact of race and gender on outcomes from patients with stage IV non-small cell lung cancer receiving immunotherapy: A retrospective study performed at Mitchell Cancer Institute, Mobile, AL.

e20562 Background: The use of immunotherapy is associated with improved survival among patients with Non-Small Cell Lung Cancer (NSCLC). The impact of race and gender on survival with immunotherapy is limited and conflicting. Our study sought to examine the influence of race and gender on the outcomes of patients with stage IV NSCLC treated with first line immunotherapy. Methods: A retrospective review of 91 patients with stage IV NSCLC was performed. Patients with confirmed NSCLC who received first line immunotherapy between January 2018 to December 2022 identified from Mitchell Cancer Institute (MCI) Cancer Registry were analyzed for clinical outcomes. Disease-specific survival was evaluated and compared among patients across racial/ethnic categories using mean, standard deviation, median, and interquartile range. Mean outcomes for numerical factors were compared using a t-test, while median outcomes were compared using a median test. The association between categorical factors was assessed using Fisher’s test, with a significance level of 0.05. Results: A total of 91 patients (45 [49.5%] male) were included in the analysis, with 61 White (67.0%; mean [SD] age, 70.2 [10.1] years) and 30 African American [AA] (32.9%; mean [SD] age, 65.9 [9.2] years). Based on histological analysis 71(77.8%) were adenocarcinoma, 18 (19.8%) squamous cell carcinoma, and 2 (2.2%) adenosquamous carcinoma. Mean progression free survival (PFS) was greater in Whites (315 vs 178 days; p = 0.041). Mean overall survival (OS) was longest in Whites (397 vs 218 days; p = 0.016). No significant differences were observed in age at diagnosis (p = 0.096), mean cycles given (p = 0.406) and PD-L1 status (p = 0.077) between Whites and AA’s. No significant differences were observed in age at diagnosis (p = 0.184), mean cycles given (p = 0.729) and time to progression (p = 0.769) or death (p = 0.938) between genders. PD-L1 distribution differed between males and females (p = 0.021) with a higher percentage of females having PD-L1 ≥1% (63% vs 38%). Conclusions: Our study showed longer PFS and OS in White patients receiving first line immunotherapy in stage IV NSCLC regardless of age at diagnosis or PD-L1 status. No differences in PFS or OS were seen between genders irrespective of age at diagnosis, mean cycles of treatment or PD-L1 status. These findings demonstrate the importance of integrating race, ethnicity, and gender as prognostic variables in the evaluation of patients with NSCLC in the incorporation of social determinants of health. This inclusion is crucial to obtain a more comprehensive understanding of how these factors influence patient outcomes.

Correlation of immune phenotypes derived from H&amp;E-stained whole slide images with prognosis and response to checkpoint inhibitors in NSCLC.

8539 Background: The classification of tumors as inflamed, excluded or desert based on spatial patterns of tumor infiltrating lymphocytes (TILs) is a potential biomarker of patients likely to respond to checkpoint inhibitors (CPI). However, the subjectivity of manual methods to assess these immune phenotypes (IPs) and poor standardization in the methods and thresholds to define IPs have hampered their clinical adoption. Here, we describe a data-driven approach to inform IP threshold selection on hematoxylin and eosin (H&amp;E)-stained whole slide images (WSI) by maximizing differences in overall survival (OS) between IPs. Methods: A model to classify the IPs of NSCLC samples from H&amp;E images was developed using PathExplore models applied to a TCGA non-small cell lung cancer (NSCLC) cohort of LUAD (N=459) and LUSC (N=424). TIL densities were extracted within cancer and stroma for 0.01 mm2 patches tiled across each WSI. Cut-offs to define cancer and stroma patches as hot or cold were defined based on the 75th and 50th percentiles, respectively, of cancer TIL (cTIL) densities (386 cTIL/mm2) and stroma TIL (sTIL) densities (423 sTIL/mm2) in a TCGA cohort of 4,082 H&amp;E-stained WSI from 10 epithelial tumor types. Hierarchical fitting yielded optimal thresholds minimizing the p-values of OS differences between IPs, leading to classifications of inflamed (iIP, &gt;40% cancer hot patches), excluded (eIP, ≤40% cancer hot patches; &gt;45% stromal hot patches) and desert (dIP, ≤40% cancer hot patches; ≤45% stromal hot patches). The model was then deployed in a clinical cohort of PD-(L)1 inhibitor-treated NSCLC patients (N=95) enrolled in the BIP precision medicine study (NCT02534649; Institut Bergonié, Bordeaux, France). Model-predicted IPs were compared to progression-free survival (PFS) and OS. FDR correction was done with Benjamini-Hochberg. Results: In the TCGA NSCLC cohort, model-predicted iIP (N=196) and eIP (N=607) patients had significantly better OS compared to dIP (N=80; HR=0.53, p=0.003 and HR=0.59, p=0.003, respectively). In the clinical cohort, cTIL density and fraction of hot epithelial patches were significantly associated with PFS (HR=0.64, q=0.04 and HR=0.69, q=0.04, respectively). PFS was significantly shorter in model-predicted eIP patients (N=46) compared to iIP (N=39; HR=0.54, p=0.045). Notably, in PD-L1(-) patients (N=43, tumor proportion score ≤1%), iIP patients had longer PFS than eIP and dIP patients (HR=0.35, p=0.02). No difference in PFS was observed for PD-L1(+) patients. Conclusions: We developed a data-driven approach for predicting IPs using patch-level TIL features. Model-predicted IPs were prognostic in a TCGA NSCLC dataset and predictive of PFS in a CPI- treated clinical NSCLC cohort. Association of IP and PFS was independent of PD-L1 status, potentially allowing the identification of PD-L1(-) patients who may derive greater benefit from CPI.

Lung-MAP S1800D: A phase II/III study of N-803 (ALT-803) plus pembrolizumab versus standard of care in participants with stage IV or recurrent non-small cell lung cancer (NSCLC) previously treated with anti-PD-1 or anti-PD-L1 therapy.

2619 Background: IL-15 is a member of the IL-2 common gamma chain family of cytokines. N-803 is IL-15 administered in complex with IL-15 receptor alpha. Lung cancer, despite advances in targeted therapies and immunotherapy, remains the leading cause of cancer related death in the United States. Strategies to improve the performance of immunotherapy in advanced NSCLC is a clinical unmet need. Methods: Lung-MAP S1800D was a randomized study comparing N-803 plus pembrolizumab (NP) to investigators’ choice standard-of-care chemotherapy (SoC) for previously treated advanced NSCLC. Patients were enrolled into an Acquired Resistance Cohort (ARC) if disease progression on prior anti-PD-(L)1 occurred &gt; 84 days from start of treatment and otherwise into a Primary Resistance Cohort (PRC). The ARC was a phase II/III study with a sample size goal of 334 patients. The PRC was a phase II with sample size of 134 patients. The first interim analysis (IA1) in the ARC evaluated futility among the first 25 patients treated with NP which required ≥1 response and ≥50% with disease control at 12 weeks to continue accrual. The primary endpoint in both cohorts was overall survival (OS). Secondary endpoints were progression-free survival (PFS), response, and toxicity. Results: Accrual in the ARC and PRC were closed at the IA1 in the ARC with 74 pts in the ARC and 8 in the PRC. Of the 74 ARC patients, 71 met eligibility (36 SoC, 35 NP), and 32 pts on each arm received treatment. With 44 events, OS was not significantly different between the two arms (HR (95%CI: 0.73(0.40-1.36), p=0.32) with a 12-month OS rate of 25% with SoC and 44% with NP. With 61 events, PFS was not significantly different but numerically worse (HR (95% CI): 1.29 (0.78-2.13, 95% CI), p=0.33). There were 3 unconfirmed partial responses and 1 confirmed complete response with NP and 3 unconfirmed partial responses and 2 confirmed partial responses with SoC. On the NP arm, there were 10 Grade 3 (1 hematologic) and 1 Grade 5 treatment-related adverse events reported as Disease Progression (34% Grade 3+ TRAE). The Grade 3+ TRAE rate on the SoC arm was 53% with 10 Grade 3+ hematologic toxicities. Conclusions: While the study failed to continue accrual past IA1, there is an indication of a subgroup that might benefit from NP with a potential OS difference at 12 months. NP was safe when compared to SoC, and responses were seen in both treatment arms, including partial and complete responses in the NP group. Evaluation of tumor and patient characteristics will be critical to define if there are those who may benefit from N-803 plus pembrolizumab. Clinical trial information: NCT05096663 .