3rd Line Research Articles

Clinicogenomic predictors of outcomes in patients with hepatocellular carcinoma treated with immunotherapy.

Immune checkpoint inhibitor (ICI) combinations extend overall survival (OS) while anti-PD-1/L1 monotherapy is non-inferior to sorafenib in treatment-naïve, patients with advanced hepatocellular carcinoma (HCC). Clinicogenomic features are posited to influence patient outcomes. The primary objective of this retrospective study was to define the clinical, pathologic, and genomic factors associated with outcomes to ICI therapy in patients with HCC. Patients with histologically confirmed advanced HCC treated with ICI at Memorial Sloan Kettering Cancer Center from 2012 to 2022 were included. Association between clinical, pathological, and genomic characteristics were assessed with univariable and multivariable Cox regression model for progression-free survival (PFS) and OS. Two-hundred and forty-two patients were treated with ICI-based therapy. Patients were predominantly male (82%) with virally mediated HCC (53%) and Child Pugh A score (70%). Median follow-up was 28 months (0.5-78.4). Median PFS for those treated in 1st line, 2nd line and ≥ 3rd line was 4.9 (range: 2.9-6.2), 3.1 (2.3-4.0), and 2.5 (2.1-4.0) months, respectively. Median OS for those treated in 1st line, 2nd line, and ≥ 3rd line was 16 (11-22), 7.5 (6.4-11), and 6.4 (4.6-26) months, respectively. Poor liver function and performance status associated with worse PFS and OS, while viral hepatitis C was associated with favorable outcome. Genetic alterations were not associated with outcomes. Clinicopathologic factors were the major determinates of outcomes for patients with advanced HCC treated with ICI. Molecular profiling did not aid in stratification of ICI outcomes. Future studies should explore alternative biomarkers such as the level of immune activation or the pretreatment composition of the immune tumor microenvironment.

The journey to LDL-C goal in the era of combination lipid lowering therapy and why it matters for ASCVD patients: a UK perspective

Objective We aim to explore the concept of distance and journey to goal, and consideration of these 2 elements a priori when choosing LLT. Methods Modelling of expected % LDL-C reductions was carried out on a range of hypothetical patients’ baseline LDL-C values prior to any LLT being commenced. Therapies were then added in a stepwise manner based on the pathway demonstrated in current national guidance and compared with goal achievement on a novel LLT optimization pathway implemented in Morecambe Bay NHS Trust. Results Modelling of a stepwise lipid management pathway shows that high-intensity statin monotherapy is not sufficient in most modelled baseline LDL-C scenarios to achieve guideline-recommended goals. Furthermore, ezetimibe second line may preclude 3rd line injectable prescribing and lead to “ezetimibe limbo” where the patient is now below the reimbursement threshold for injectable prescribing but still not achieving their LDL-C target. Overall goal achievement is poor across the spectrum of modelled LDL-C levels. In contrast, by following the Morecambe Bay pathway all patients on statin for the range of hypothetical baseline LDL-C levels can reach an LDL-C target of < 1.8 mmol/L. Conclusions This study identifies a therapeutic gap when following a stepwise approach highlighted by recent national guidance. Our proposal of a novel pathway highlights that the order in which drugs are added is important in the context of national reimbursement thresholds and allows LDL-C goal to be reached in a timely manner, regardless of the starting baseline LDL-C level.

Efficacy of anti PD-1 therapy in children and adolescent melanoma patients (MELCAYA study)

BackgroundData on the efficacy and safety of anti PD-1 antibodies in children and adolescents (CA) with melanoma are lacking. The aim of this study was to determine outcomes of CA melanoma patients receiving anti PD-1 antibodies. MethodsMelanoma patients ≤18 years treated with anti PD-1 were retrospectively retrieved from 15 academic centers. Information on histopathological diagnosis, surgical treatment, systemic therapy, objective response rate (ORR), safety profile was collected. Progression-free survival (PFS) and overall survival (OS) were assessed by Kaplan-Meier method. ResultsBetween April 2016 and March 2024, 99 patients treated with systemic therapy were retrieved, 81 treated with anti PD-1 therapy. Median age was 14 years (range 2–18 years), 37 pts were ≤12 yrs. Overall, 38 CA patients received anti PD-1 in adjuvant setting, and the 3-year PFS and OS were 70.6 % and 81.1 %, respectively. Two patients received anti-PD-1 based neoadjuvant treatment, both had a pathologic complete response and remain disease free. Fifty-six received a systemic therapy for advanced disease and among them, 43 received anti PD-1-based therapy for advanced disease in 1st line, while 12 and 5 pts received a 2nd and 3rd line, respectively. Among patients receiving a 1st line therapy with anti PD-1 monotherapy the ORR was 25 %, and the 3-year OS was 34 %. Toxicities were consistent with previous studies in adult melanoma patients. ConclusionsOur study provides the first evidence of efficacy of anti PD-1 in CA melanoma patients and supports the use of anti PD-1 therapy in pts ≤18 years, included those <12 years.

Rationale and Design of the COPERNIC Trial: A Study of On-treatment ctDNA Changes in Chemo-refractory Colorectal Cancer Patients

BackgroundEvidence suggests that ctDNA may be a reliable biomarker to monitor metastatic colorectal cancer (CRC) evolution. Nevertheless, evidence on the potential of liquid biopsy in this setting is still low quality, mostly consisting of retrospective studies. MethodsCOPERNIC is an international, multicenter clinical trial. The pilot study aims to confirm the predictive potential of early on-treatment ctDNA dynamics, and inform the design of a larger ctDNA-driven trial. Advanced CRC patients who are candidates for ≥3rd lines of systemic therapy undergo longitudinal blood sample collection during treatment (day 1, 15 and 29 for 2- or 4-weekly treatment regimens; day 1, 22 and 43 for 3-weekly treatment regimens) and at each imaging assessment. ctDNA analyses are carried out with the FoundationOne Liquid CDx and FoundationOneMonitor assays, and ctDNA changes during treatment are correlated with radiologic response (as assessed every 8-12 weeks by RECIST v1.1). The primary objective is to select the optimal timepoint and cut-off value for early ctDNA changes (at day 15/22) to predict progressive disease as best radiological response with a high positive predictive value. The cut-off value for ctDNA will be defined based on nonparametric ROC-curves with bootstrapping. Based on the expected rate of progressive disease and statistical assumptions, 109 patients are needed to be screened to have 87 assessable patients. COPERNIC is sponsored by the Institut Jules Bordet, and supported by Roche and Foundation Medicine. Recruitment is open in 13 centres across Belgium and France. The study is registered with clinicaltrials.gov (NCT05487248).

35 Efficacy of treatments post-lenvatinib in patients with advanced renal cell carcinoma (aRCC)

Abstract Background Lenvatinib is a tyrosine kinase inhibitor (TKI) that targets both vascular endothelial growth factor (VEGF) receptor and fibroblast growth factor receptor. It has demonstrated efficacy both in the upfront and refractory disease settings. However, there is a lack of data surrounding the efficacy of TKIs post-lenvatinib exposure. In this study, we investigate the activity of therapies post-lenvatinib in patients with aRCC. Methods We conducted a retrospective analysis utilizing the International Metastatic Database Consortium (IMDC). Patients having received treatment post-lenvatinib exposure were eligible and divided into two cohort: patients post-1st line lenvatinib (2nd line cohort) and patients post-2nd line lenvatinib (3rd line cohort). The primary objective was objective response rate (ORR) and time to treatment failure (TTF). ORR was summarized with 95% two-sided exact binomial confidence interval. TTF was defined as time from treatment initiation to drug cessation for any reason censored at the date of last follow-up. Results Overall, 84 patients received 1st line lenvatinib of whom 43 (51%) remain on therapy, 20 (24%) received 2nd line treatment, and 21 (25%) received no subsequent treatment. The median duration of prior lenvatinib was 9.7 months. All patients received 1st line pembrolizumab + lenvatinib (ORR 50%, median TTF 19.7 months). Reason for lenvatinib discontinuation was progression (50%), progression + toxicity (20%), toxicity (15%), or other (15%). For the 2nd line cohort, median age was 61 years, most patients were male (85%), had prior nephrectomy (75%), clear cell histology (85%), and were IMDC intermediate/poor risk (55%). 2nd line therapy regimens included TKI monotherapy (80%), TKI-IO (5%), and other (15%). The median follow up from 2nd-line treatment initiation was 4.9 months. The ORR to 2nd line treatment was 5% (95% CI 0.2-25) and median TTF was 5.8 months (95% CI 1.9-14.9). Of 2nd line lenvatinib-exposed patients (n=84), 24 (29%) remain on treatment, 34 (40%) received 3rd line treatment, and 26 (31%) did not receive additional therapy. The median duration of prior lenvatinib was 5.9 months. Most patients received 2nd line everolimus + lenvatinib (97%) (ORR 31%, median TTF 9.2 months). Reason for lenvatinib discontinuation was progression (59%), progression + toxicity (9%), toxicity (12%), or other (21%). For the 3rd line cohort, median age was 67 years, most patients were male (68%), had prior nephrectomy (88%), clear cell histology (68%), and were IMDC intermediate/poor risk (77%). 3rd line treatments included TKI alone (50%), IO-TKI (38%), and other (12%). The median follow up from 3rd-line treatment initiation was 14.9 months. The ORR to 3rd line treatment was 12% (95% CI 3.3-27) and median TTF was 2.8 months (95% CI 1.9-7.4). Conclusions In this analysis, we demonstrate modest activity of TKI-based therapy post-lenvatinib exposure. Our study highlights the need for improved treatment options for patients progressing on lenvatinib-based therapies.

Socio-economic burden of chronic graft-versus-host disease after allogenic bone marrow transplantation

Aim. Chronic graft-versus-host disease (crGVHD) socio-economic burden assessment in the Russian Federation.Materials and methods. The assessment was performed in children aged 12–18 years and adults aged over 18 years. The payer's position is "Government". The modeling horizon is 5 years. Direct medical (costs of crGVHD drug therapy; prevention of complications of crGVHD therapy and their treatment in case of occurrence) are taken into account; the cost of visits with the symptoms of crGVHD), as well as direct non-medical costs and indirect costs due to the patient’s disability or disability of their relatives.Results. The total costs for the population of patients with crGVHD aged 12–18 years ranged from 498,120,490 rubles to 1,487,918,739 rubles, in adults — from 969,019,644 rubles to 3,290,898,722 rubles from a five-year perspective, depending on the simulated population size and the choice of the 3rd line therapy regimen. A one-sided sensitivity analysis demonstrated that the main changes in the size of the overall economic burden were observed in the case of changes in the size of the target population, as well as in the cost of therapy.Conclusions. The conducted clinical and economic analyses demonstrated that significant losses of both the health system and the state as a whole were associated with crGVHD after allogeneic bone marrow transplantation, which requires the development of new prevention strategies.

Matching-Adjusted Indirect Comparisons of Axicabtagene Ciloleucel to Mosunetuzumab for the Treatment of Relapsed/Refractory Follicular Lymphoma

Axicabtagene ciloleucel (axi-cel) was the first chimeric antigen receptor (CAR) T-cell therapy approved for relapsed/refractory (R/R) follicular lymphoma (FL) patients, while mosunetuzumab was the first bispecific monoclonal antibody approved in this population. In the absence of head-to-head evidence, this study sought to conduct a matching-adjusted indirect comparison (MAIC) to estimate the comparative efficacy and safety of these treatments in 3rd line or higher (3L+) FL. The evidence base consisted of individual patient data (IPD) of all enrolled patients, regardless of infusion status, from the single-arm axi-cel trial, ZUMA-5 (NCT03105336), and aggregate data from the mosunetuzumab FL trial (NCT02500407) from publications identified through a systematic review. Efficacy outcomes were progression-free survival (PFS), duration of response (DoR), objective response rate (ORR), complete response rate (CRR). Analyses used independent central review for both trials, where possible. Safety outcomes were cytokine release syndrome (CRS), neurological events (NE), and treatment-related adverse events (TRAEs). Unanchored MAICs were conducted to align ZUMA-5 to the patient characteristics of the mosunetuzumab trial. For each outcome, prognostic factors were identified a priori through quantitative analysis and clinical experts. For time-to-event outcomes, hazard ratios (HRs) were estimated using Cox regression using IPD from ZUMA-5 and pseudo-IPD extracted from Kaplan-Meier plots for mosunetuzumab. Patient characteristics were well-aligned between trials leading to large effective-sample sizes after matching, ranging from 93.4 to 115.5, for ZUMA-5 (n = 127). In comparisons to mosunetuzumab (n = 90), axi-cel was associated with improved PFS (HR: 0.39; 95% confidence interval [CI]: 0.24-0.62) and DoR (HR: 0.45; 95% CI: 0.27-0.76). Similarly, axi-cel led to higher ORR (OR: 3.87; 95% CI: 1.53-9.76) and CRR (OR: 2.80; 95% CI: 1.50-5.26). Although axi-cel was associated with a higher rate of all-grade CRS (OR: 5.54; 95% CI: 2.63-8.94) and NEs (OR: 3.54; 95% CI: 1.28-9.83), differences in grade ≥3 CRS and TRAEs were not statistically significant. Findings from this study show improved efficacy and more durable response for the treatment of 3L+ R/R FL with axi-cel relative to mosunetuzumab, with increased odds of all-grade CRS and NE, but not G3+ CRS and TRAEs.

Comprehensive genomic profiling of ESR1, PIK3CA, AKT1, and PTEN in HR(+)HER2(−) metastatic breast cancer: prevalence along treatment course and predictive value for endocrine therapy resistance in real-world practice

BackgroundThe treatment landscape for HR(+)HER2(−) metastatic breast cancer (MBC) is evolving for patients with ESR1 mutations (mut) and PI3K/AKT pathway genomic alterations (GA). We sought to inform clinical utility for comprehensive genomic profiling (CGP) using tissue (TBx) and liquid biopsies (LBx) in HR(+)HER2(−) MBC.MethodsRecords from a de-identified breast cancer clinicogenomic database for patients who underwent TBx/LBx testing at Foundation Medicine during routine clinical care at ~ 280 US cancer clinics between 01/2011 and 09/2023 were assessed. GA prevalence [ESR1mut, PIK3CAmut, AKT1mut, PTENmut, and PTEN homozygous copy loss (PTENloss)] were calculated in TBx and LBx [stratified by ctDNA tumor fraction (TF)] during the first three lines of therapy. Real-world progression-free survival (rwPFS) and overall survival (rwOS) were compared between groups by Cox models adjusted for prognostic factors.Results ~ 60% of cases harbored 1 + GA in 1st-line TBx (1266/2154) or LBx TF ≥ 1% (80/126) and 26.5% (43/162) in LBx TF < 1%. ESR1mut was found in 8.1% TBx, 17.5% LBx TF ≥ 1%, and 4.9% LBx TF < 1% in 1st line, increasing to 59% in 3rd line (LBx TF ≥ 1%). PTENloss was detected at higher rates in TBx (4.3%) than LBx (1% in TF ≥ 1%). Patients receiving 1st-line aromatase inhibitor + CDK4/6 inhibitor (n = 573) with ESR1mut had less favorable rwPFS and rwOS versus ESR1 wild-type; no differences were observed for fulvestrant + CDK4/6 inhibitor (n = 348).ConclusionOur study suggests obtaining TBx for CGP at time of de novo/recurrent diagnosis, followed by LBx for detecting acquired GA in 2nd + lines. Reflex TBx should be considered when ctDNA TF < 1%.

Efficacy of trifluridine/tipiracil in 3rd or 4th line of treatment in patients with advanced or metastatic colorectal cancer: A real-world data analysis with subgroups exploration.

e15576 Background: Patients with unresectable mCRC can achieve long-term overall survival (OS) thanks to modern systemic treatment and subsequent lines of therapy. With effective 1st, 2nd and next lines of therapy, their condition often allows further lines of treatment. Since publication of RECOURSE study, one of the most frequently used drugs in the 3rd or 4th line is trifluridine/tipiracyl (FTD/TPI). There is limited data on optimal selection of patients in order to obtain the greatest therapeutic gain with the least toxicity. The aim of this study was to find and define groups of patients who would benefit most from this treatment in a real-world mCRC patient population. Methods: Consecutive patients with unresectable or metastatic colorectal cancer (mCRC) started treatment with FTD/TPI between 1/Sep/2019 and 31/Dec/2023. Clinical factors including age, gender, primary location of colorectal cancer, BRAF, RAS and MSI-H status, ECOG performance status, baseline CEA level, location of metastases, response to treatment and AEs occurrence were analyzed. Survival analyses were performed using the Kaplan-Meier method, Log-rank and chi-square tests were used for comparison between groups. Data cut-off was 31/Jan/2024. Results: 822 medical records of patients with mCRC were analyzed. In total 261 patients treated with FTD/TPI were enrolled. 204 (78%) patients were treated with FTD/TPI in 3rd line (L3) and 57 (22%) in the 4th line (L4). Patients with RASmt were statistically more numerous in the L3 group; no other statistically significant differences in baseline characteristics was found between L3 and L4 groups. Estimated median OS (mOS) was 10.2 and 14.4 months in L3 and L4 group, respectively; no statistical significance was found between them (p = 0.0866; HR = 1.34, Cl 95% 0.9-1.9). Estimated median PFS (mPFS) was 7.4 and 4.6 months in L3 and L4 group, respectively, and statistically significant differences in mPFS was found in the L3 group in comparison to L4 (p = 0.0079; HR = 0.65, Cl 95% 0.5-0.9). Disease progression (PD) was found in 187 (92%) and 50 (88%) patients in the L3 and L4 group, respectively. PD was most common reason for discontinuing therapy. Most frequent serious AEs of grade 3 or higher were diarrhea, neutropenia and weakness. There were no grade 5 AEs and toxicity rarely led to discontinuation of FTD/TPI therapy. 102 (39%) patients were qualified for next line sequential systemic treatment after L3 or L4. Conclusions: Patients with mCRC who were treated with FTD/TPI in the 3rd line had better PFS than patients treated in 4th line, but mOS was statistically non-significantly better in L4 than L3, which requires further analysis. FTD/TPI is an effective therapy in subsequent lines of treatment for mCRC patients with acceptable toxicity.

Genomic alterations (GA) in ESR1, PIK3CA, AKT1, and PTEN in HR(+)HER2(-) patients (pts) with metastatic breast cancer (MBC): Co-occurrence and prevalence along treatment course.

1060 Background: The standard of care for HR(+) HER2(-) MBC is evolving with new biomarker-guided drugs, including novel selective estrogen receptor degraders (SERDs) for pts with ESR1 mutations (mut) and PIK3CA and AKT inhibitors for pts with alterations in the PI3K pathway (PIK3CA, AKT1, and PTEN). This study aimed to evaluate the co-occurrence and the prevalence of these GA in different metastatic lines of therapy in real-world practice for MBC. Methods: This study included pts with HR(+) HER2(-) MBC who underwent genomic testing using Foundation Medicine tissue or liquid comprehensive genomic profiling assays, with specimens collected within 60 days of therapy initiation. Clinical data were obtained by the nationwide (US-based) de-identified Flatiron Health and Foundation Medicine real-world clinicogenomic breast cancer database (FH-FMI CGDB), originated from ~280 US cancer clinics (~800 sites of care) between 01/2011-06/2023. The prevalence of ESR1mut, PIK3CAmut, AKT1, PTENmut, and PTEN copy loss were calculated in tissue (TBx) and liquid biopsies (LBx) collected at time of 1st, 2nd, and 3rd lines of therapy. The co-occurrence of GA in TBx was evaluated. Results: Among patients with a TBx collected around therapy initiation (n=2,536), 81.5% (2,066) was collected at 1st, 10.2% (258) at 2nd, and 8.3% (212) at 3rd line. Among patients with a LBx collected around therapy initiation (n=545), 47% (256) was collected at 1st, 33.2% (181) at 2nd, and 19.8% (108) at 3rd line. In TBx, the prevalence of GA detected in 1st, 2nd, and 3rd lines were 58.6% (1,210), 67.4% (174), and 71.2% (151), respectively. In LBx, the prevalence was 42.7% (121), 65.7% (119), and 60.1% (65), respectively. Among LBx with tumor fraction ³1%, the prevalence was 67.9% (74/109), 89.1% (81/91), and 84.3% (43/51). Among patients with a GA detected in TBx 1st (n=1,210), 2nd (n=174), and 3rd (n=151) lines, both ESR1mut and a PIK3 pathway GA was detected in 7.4% (90), 12.1% (21), and 23.8% (36), respectively. Conclusions: In our dataset, 42-59% of patients with HR(+) HER2(-) MBC have at least one GA detected by TBx or LBx in 1st line, with prevalence increasing in later lines of therapy to 60-71% due to acquisition of an ESR1mut. PTEN copy loss is detected at much higher rates in TBx than LBx, consistent with the known inherent limitations of ctDNA testing. Serial genomic testing with LBx with reflex TBx if alterations are not identified should strongly be considered with the recent approvals of several molecularly targeted agents for HR(+) HER2(-) MBC. [Table: see text]

Efficacy of post-lenvatinib treatments in patients (pts) with advanced renal cell carcinoma (aRCC).

4538 Background: Lenvatinib is a tyrosine kinase inhibitor (TKI) that targets both vascular endothelial growth factor (VEGF) receptor and fibroblast growth factor receptor. It has demonstrated efficacy both in the upfront and refractory disease settings. However, there is a lack of data surrounding the efficacy of TKIs post-lenvatinib exposure. In this study, we investigate the activity of post-lenvatinib therapies in pts with aRCC. Methods: We conducted a retrospective analysis utilizing the International Metastatic Database Consortium (IMDC). Pts having received treatment post lenvatinib exposure were eligible and divided into two cohort: pts post-1st line lenvatinib (2nd line cohort) and pts post-2nd line lenvatinib (3rd line cohort). The primary objective was objective response rate (ORR) and time to treatment failure (TTF). ORR was summarized with 95% two-sided exact binomial confidence interval. TTF was defined as time from treatment initiation to drug cessation for any reason censored at the date of last follow-up. Results: Overall, 84 pts received 1st line lenvatinib of whom 43 (51%) remain on therapy, 20 (24%) received 2nd line treatment, and 21 (25%) received no subsequent treatment. All pts received 1st line pembrolizumab + lenvatinib (ORR 50%, median TTF 9.7 months). Reason for lenvatinib discontinuation was progression (50%), progression + toxicity (20%), toxicity (15%), or other (15%). For the 2nd line cohort, median age was 61 years, most pts were male (85%), had prior nephrectomy (75%), clear cell histology (85%), and were IMDC intermediate/poor risk (55%). 2nd line therapy regimens included TKI monotherapy (80%), TKI-IO (5%), and other (15%). The ORR to 2nd line treatment was 5% (95% CI 0.2-25) and median TTF was 5.8 months (95% CI 1.9-14.9). Of 2nd line lenvatinib-exposed pts (n=84), 24 (29%) remain on treatment, 34 (40%) received 3rd line treatment, and 26 (31%) did not receive additional therapy. Most pts received 2nd line everolimus + lenvatinib (97%) (ORR 39%, median TTF 5.9 months). Reason for lenvatinib discontinuation was progression (59%), progression + toxicity (9%), toxicity (12%), or other (21%). For the 3rd line cohort, median age was 67 years, most pts were male (68%), had prior nephrectomy (88%), clear cell histology (68%), and were IMDC intermediate/poor risk (77%). 3rd line treatments included TKI alone (50%), IO-TKI (38%), and other (12%). The ORR to 3rd line treatment was 12% (95% CI 3.3-27) and median TTF was 2.8 months (95% CI 1.9-7.4). Conclusions: In this analysis, we demonstrate modest activity of TKI-based therapy post-lenvatinib exposure. Our study highlights the need for improved treatment options for pts progressing on lenvatinib-based therapies.

Outcomes with trastuzumab deruxtecan (T-DXd) by HER2 status and line of treatment in a large real-world database of patients with metastatic breast cancer.

1077 Background: In the clinical trial setting, T-DXd improves survival for patients with HER2+ and HER2-low metastatic breast cancer (MBC), and has also shown preliminary activity in HER2-0 MBC. Little real-world evidence is available on the performance of T-DXd. Methods: We conducted a retrospective observational study using the nationwide Flatiron Health electronic health record (EHR)-derived deidentified database to evaluate the real-world activity of T-DXd. We included patients with MBC who initiated T-DXd between 12/2019 and 1/2023. Tumors were categorized as HER2+ if positive at any timepoint, with HER2- cases further divided into HER2-low (IHC 1+ or 2+ not amplified) and HER2-0 (IHC 0) according to the last biopsy before T-DXd. Real world progression-free survival (rwPFS) and overall survival (OS) were estimated using the Kaplan-Meier method. Results: A total of 930 patients were included: 636 with HER2+, 268 with HER2-low and 26 with HER2-0 MBC. Among HER2- patients, 241 (82.0%) were HR+ and 53 (18.0%) were HR- (i.e., triple-negative). A total of 572 (61.5%) patients were White, 101 (10.9%) Black/African American, 88 (9.4%) Hispanic/Latino, 37 (4.0%) Asian and 132 (14.2%) other or missing. Median age was 59 [range 25, 84] and patients received T-DXd in the following lines of therapy (LOT): 54 (5.8%) as 1st line, 110 (11.8%) as 2nd line, 179 (19.2%) as 3rd line, 182 (19.6%) as 4th line and 405 (43.6%) as ≥5th line. Median number of LOT for advanced disease prior to T-DXd were 3 and 4, respectively, for HER2+ and HER2- patients. In patients with HER2+ MBC, rwPFS and OS were 12.1 and 26 months after initiating T-DXd, respectively. The longest rwPFS, 14.9 months, was observed in the 1st or 2nd LOT, followed by 14.1 months in LOT 3, 12.4 months in LOT 4, and 10.8 months in LOT ≥5. In patients with HER2- MBC, median rwPFS and OS were 6.3 and 15.5 months, respectively. The longest median rwPFS of 8.1 months was observed in the 1st or 2nd LOTs, followed by 6.9 months in LOT 4, 6.2 months in LOTs ≥5, and 6.0 months in LOT 3. Outcomes with T-DXd by HER2 and HR status are provided (Table). Updated data with a larger sample size will be presented. Conclusions: In the largest dataset to date, T-DXd showed favorable real-world activity for treating MBC, although rwPFS appeared shorter than what observed in clinical trials. Encouraging rwPFS was observed in understudied groups, such as patients with triple-negative and HR+/HER2-0 MBC. [Table: see text]

A phase 2a study of NT-I7 (efineptakin alfa), a long-acting IL-7, and pembrolizumab to evaluate efficacy, including overall survival, in hard-to-treat gastrointestinal tumors.

2621 Background: Microsatellite-stable colorectal (MSS-CRC) and pancreatic cancer (PDAC) are immunologically cold tumors with null response to checkpoint inhibitors (CPI). NT-I7, a long-acting IL-7, in combination with pembrolizumab (pembro) has shown to significantly increase intratumoral T cell infiltration and elicit some tumor control in these hard-to-treat gastrointestinal indications. The original MSS-CRC and PDAC cohorts, enrolling 25 patients (pts) each, were expanded to 50 pts per indication. Here, we provide an updated analysis including the original and expansion cohorts. Methods: Open-label Phase 2a study in subjects with relapsed/refractory CPI-naïve MSS-CRC and PDAC; NT-I7 1200 µg/kg IM every 6 weeks (Q6W), pembro 200 mg IV Q3W. Antitumor activity assessed by RECIST v1.1/iRECIST. Results: As of 02OCT2023, 98 subjects were efficacy-evaluable (50 MSS-CRC, 48 PDAC); median age 61.0 years, all with baseline ECOG status 0-1. 95.9% (94/98) received study treatment ≥3rd line, with a median follow -up of 6.1 months. Safety profile for NT-I7 was similar to previous reports. In MSS-CRC pts, 1 pt and 3 pts achieved confirmed partial responses per RECIST and iRECIST, respectively; median duration of response (DOR/iDOR) was 13.1/13.0 months (mo) and 2 responders were alive with no progression at data cutoff. Disease control rate (DCR/iDCR) was 36.0% (18/50) / 38.0% (19/50), with a duration of response and stable disease (DORSD/iDORSD) of 3.2/14.5mo. Interestingly, iDCR was 71.4% (5/7) in MSS-CRC pts with primary tumor in the rectum, versus 30.8% (12/39) in pts with primary tumor in the colon. With 11 pts on follow up, progression free survival (PFS/iPFS) was 1.5/3.8 mo. Among PDAC pts, 2 and 3 pts achieved confirmed PR and iPR, with DOR/iDOR of NE/9.7 mo; 1 responder was alive with no progression at data cutoff. DCR/iDCR was 25.0% (12/48) / 27.1% (13/48), and DORSD/iDORSD was 2.9/9.8mo. With 11 pts on follow up, PFS/iPFS was 1.4/2.1 mo. Clinical benefit in MSS-CRC and PDAC was most evident by a notable median overall survival (mOS); 13.2 mo in MSS-CRC [95% CI 8.9 – 18.6 mo] and 11.1 mo in PDAC [95% CI 4.1 – 13.3 mo], compared to historical mOS for standard of care, 10.8 mo (MSS-CRC) and 6.1 mo (PDAC). Conclusions: NT-I7 and pembro treatment was associated with longer mOS relative to historical studies. Identification of predictive biomarkers that may define pts with higher likelihood of clinical benefit would be a promising step to maximizing potential of the NT-I7 + pembro treatment combination for pts with these indications. Research in this area is ongoing. Clinical trial information: NCT04332653 .

Systemic therapy in NF-1 associated malignant peripheral nerve sheath tumors (MPNST).

11583 Background: MPNST are rare sarcomas that have a poor prognosis and are challenging to treat. Around 50% of MPNSTs are associated with NF1, 40% are sporadic, and 10% are radiation-associated. Doxorubicin (D) based chemotherapy (CT) regimens are commonly used in the front-line, and combinations have better response rates than single agents. Response to second-line regimens is not well documented. The objective of this study is to document efficacy of systemic regimens in MPNST as a benchmark for future novel therapeutics in this space. Methods: 72 patients (pts) with NF1-associated MPNSTs in an MD Anderson database were reviewed retrospectively from EMR. Variables collected included demographics, primary MPNST location, localized vs metastatic, and treatments received (+/- radiation, +/-surgery, +/- lines of CT). Responses (progressive disease (PD), stable disease (SD), partial response (PR)) as interpreted by treating physician and radiology reports and time on treatment for 1st and 2nd line chemotherapy regimens were assessed. Cox proportional-hazard regression was used to investigate associations of variables with overall survival (OS: from time of diagnosis to death or last follow-up). Results: Demographic and CT details are provided (Table). The median age at diagnosis was 34.5 yrs (range: 9-71). The most common MPNST location was the trunk (56.94%). D based regimens were the most frequent 1st-line CT (65.7%), and gemcitabine- docetaxel was the most frequent 2nd-line CT (33.3%). In pts with metastases, 74.4% got 1st line, 44.7% 2nd line, 25.5% 3rd line, and 12.7% got 4th line therapy. Median follow-up was 2.46 yrs (IQR 1.24-9.52). The median time on treatment for 1st and 2nd line CT was 110 (IQR 48-153) and 57 (IQR 41-133) days, respectively. In the multivariate COX model for the entire cohort, male sex (HR: 0.35, 95%CI: 0.18-0.68), lower extremity primary (HR: 0.09, 95%CI: 0.02-0.40), and absence of metastases at last follow up (HR: 0.05, 95%CI: 0.02-0.17) were associated with decreased risk of death. Additional analysis on efficacy and survival of the systemic therapy regimens will be presented at the meeting. Conclusions: In this study &lt; 50% pts with advanced NF1 related MPNST made it to 2nd line CT and outcomes were inferior to the predominantly D based 1st line regimens. This highlights an area of high unmet need. [Table: see text]

Interchangeability of anti-PD1 antibodies in patients (pts) with metastatic NSCLC.

e20553 Background: PD-(L)1 inhibitors are approved immunotherapy agents in treatment for mNSCLC, however, limited data exists about switching between different anti-PD1 antibodies. Our study aimed to evaluate the safety and efficacy of the interchangeability of nivolumab, pembrolizumab, and atezolizumab in patients (pts) with mNSCLC in the 2nd and 3rd line. Methods: We retrospectively evaluated data of pts with mNSCLC treated in four oncology centers in Moscow and SaintPetersburg (Russia). Inclusion criteria were as follows: histologically confirmed mNSCLC with no driver mutations, performance status ECOG 0-2, monotherapy of anti-PD(L)1 antibody, and switching to another PD(L)1 inhibitor for any reason except progression disease or toxicity after initiating of immunotherapy. The primary end-point was progression-free survival (PFS), and secondary end-points were the rate of all grades and grade 3-4 adverse events (CTC AE 5.0). A sample size of 83 achieves 80% power to detect an equivalence difference of 15% in 3-months PFS after the change of anti-PD1 antibody with a one-sided significance level of 0,05. Results: A total of 103 patients (27% female) were included with a mean age of 66 years (33-93). 0-1, 2, and ≥3 organs with metastases have 70 (68%), 17 (17%), and 16 (16%) pts, respectively. Performance status ECOG 0, 1, and 2 was in 16 (16%), 70 (68%), and 17 (17%) pts, respectively. 50 (48%) patients received pembrolizumab, 40 (39%) pts received nivolumab, and 13 (13%) pts – atezolizumab. 48 (47%) pts were switched to atezolizumab, 28 (27%) pts were switched to nivolumab, 26 (25%) pts – to pembrolizumab. The median duration of treatment with the first anti-PD(L)1 antibody was 6 months (1-53 months), with the second PD(L)1-inhibitor was 3 months (1-41 months). Within 3 months after 1 replacement of the drug, there were 77 (74,7%) pts without progression. With the median follow-up of 8 months (1-41), the estimated 3-month PFS was 79,3%. The efficacy of switching was not associated with the agent (HR 0,97, 95CI 0,68-1,38, p = 0,9). There were 2 cases of AEs of 3-4 grades when switching to the second drug. Conclusions: Switching from one anti-PD(L)1 antibody to another in pts with mNSCLC is possible in case of limited access to PD(L)1 inhibitors.

Efficacy and toxicity of KIT targeted therapy in elderly patients with melanoma.

e23269 Background: Most cases of malignant melanoma are cutaneous, with ≈50% having a BRAF driver mutation that predicts response to MAPK pathway inhibitors. A small minority of melanoma cases harbor an activating KIT mutation, most notably in mucosal and acral melanomas (15-20%) and occasionally in those arising from chronic sun-damaged skin (2-3%). Only about a third of KIT mutated melanomas respond to targeted therapy with Imatinib, yet data is scarce and limited, especially in older patients. We sought to characterize the efficacy and toxicity of Imatinib in KIT-mutated melanoma patients over 65. Methods: We identified electronic medical records of patients over 65 that were treated with Imatinib for KIT mutated melanoma at our institution over the years 2017-2023. Data was retrospectively collected including demographics, disease characteristics, response patterns and toxicity. Results: 10 patients were identified, 50% were males. The median age at treatment initiation was 75 years (71-100). 3 patients had cutaneous primary, 3 had Vulvar/vaginal primary, 3 had acral primary and 1 had anal canal primary. All patients had normal Lactate dehydrogenase at treatment initiation and ECOG performance status was 0-2 (median 0). 9 patients received Imatinib for stage IV disease with an overall response rate (ORR) of 55% (44% partial response, 11% complete response). Median duration of response was 4 months. Responses were seen in acral melanoma (67%), vulvar/vaginal melanoma (67%) and cutaneous melanoma (33%). Median Progression free survival was 4.3 months (1.8-44), median overall survival from treatment initiation was 22.6 months (4-64). 6 patients were treated as 2nd line with an ORR of 67% and 3 were treated as 3rd line with an ORR of 33%. There was one patient with M1d disease who did not respond to treatment either systematically or intracranially. 6 patients had Exon 11 mutations (V560D/G/E, N564S and L576P) of which 4 responded (67%). 2 had Exon 17 mutations (N822C/K) of which 1 responded (50%) and the 2 others had an Exon 14 (F681I) and Exon 13 (K642E) mutation that did not respond to treatment at all. Toxicity was moderate with 90% of patients experiencing any grade AE, yet only 1 experienced toxicity requiring dose interruption with two Grade 3 toxicities - Pneumonitis and decreased neutrophil count, leading to temporary treatment interruption, dose reduction and a course of corticosteroids. The rest of the patients experienced only Grade 1-2 toxicity. 4 patients needed dose reductions, there were no treatment discontinuations due to toxicity. Conclusions: Treatment with Imatinib for elderly KIT mutated melanoma patients is safe and well tolerated with no treatment discontinuations recorded. The response rate is encouraging yet seems to vary according to the KIT mutation present as previously reported. The duration of response is usually short-lived and alternative/combination treatments are still needed for this patient population.

End of life treatment in gynecological cancers in an integrated care model: A single institution experience from Oman.

e24036 Background: The use of chemotherapy closer to the end of life (EOL) and the optimal time of initiating palliative care interventions are subjects of ongoing debate. In a traditional care model (TCM), the institution of EOL care is a complex decision-making process, that involves patient, family members, and the oncologist, and involves transfer of patients from one to the other service. Integrated care model (ICM) is a newer concept, is designed around the needs of the patients, and may have the potential to improve quality of life and reduce the cost of healthcare towards EOL. At our center, we work in ICM in organ-based clinical programs and report the use of systemic treatment in patients with gynecological cancers (GC) nearing EOL. Methods: We analyzed retrospectively the use of chemotherapy and other systemic cancer-directed therapies in patients with GC, who died between Sept 2021 and December 2023 at a single institution. Information on the cancer-type, intention of treatment, and the time between last treatment and death was obtained from electronic patient records (EPR). Results: Over the study period, a total of 22 patients passed away. All patients received palliative care as part of the integrated care. All patients had been declared ‘not for resuscitation’ before death. Twelve patients had ovarian cancer, 5 had endometrial cancer, 4 had cancer of cervix, and one had vulvar cancer. The median age at diagnosis was 53 years (29-80). The systemic treatment was 1st line in 5 patients (23%), 2nd line in 8 patients (36%), and 3rd or more line in 6 patients (27%) respectively. Three patients (14%) did not receive any cancer-directed treatment (aged 80, 71 and 61). The ultimate line of treatment was cytotoxic chemotherapy for all patients, except for one who received Olaparib, and one who received trastuzumab with chemotherapy. There was no treatment-related mortality. The median time from last treatment to death was 55 days. Of the 19 patients who received treatment, 6 patients (32%) received treatment in the last 30 days, 5 between 30 and 60 days (26%), 5 between 60 and 90 days (26%), and 3 more than 90 days before death. Four patients died at home. Age at diagnosis (51 vs 55 years), age at death (51.5 vs 57.5 years), or time from diagnosis to death (11.5 vs 22 months), did not predict whether patients received treatment within the 30 days of death or earlier. Conclusions: All patients received palliative care as part of integrated cancer care. The ICM may have led to lower use of cancer treatment near EOL. The data may provide a benchmark for the region.

The analysis of prescribing CDK4/6 inhibitors for patients with breast cancer in real clinical practice.

e13082 Background: The analysis of prescribing CDK4/6 inhibitors for patients with HR+/Her2neu- breast cancer was conducted at the St. Petersburg City Clinical Oncology Dispensary from April 2020 to October 2023. Patients in group 1 (n=365) received Palbociclib, and patients in group 2 (n=450) received Ribociclib. Methods: 99.45% of participants in the 1 group were female. In the 2 group all patients female. The average age of patients in both groups was 66 years. Results: In the first line of therapy, the clinical oncologist's choice more often leaned towards Ribociclib (p=0.0293). In later lines, Palbociclib was more frequently prescribed (p=0.0086). Ribociclib was significantly more often prescribed compared to Palbociclib in cases of primary hormone resistance (19.8%vs.13.9% (p=0.0288)). A higher "tumor burden" was documented at the start of therapy in patients receiving Ribociclib - 9.32% (5 or more loci of metastatic involvement). The frequency of overall response was significantly higher with Palbociclib (24.3%vs.11.7% (p=0.00001)), which is likely related to the higher incidence of primary hormone resistance and prescription in cases with a higher "tumor burden" in the Ribociclib group. The presence of PIK3CA mutations in patients of the Palbociclib group signifies a significant difference in response to therapy. The mPFS is 26(95% CI 16-29) when the mutation is present and 11(95% CI 9-14) when it is absent (p=0.0016). In the presence of metastatic liver impairment, significant differences in the mPFS assessment were in favor of Ribociclib (12 (95% CI 10-19) vs 11.0 (95% CI 9-15) (p&lt;0.0001)). For the presence of a BRCA mutation, age, presence of bone metastases, brain metastases, visceral metastases (excluding liver), hormonal partner, there were no statistically significant differences between the groups for progression-free survival. The mPFS in both groups was comparable, and we didn't obtain statistically significant differences. In the 1st line the mPFS is not yet reached (p=0.602). In the 2nd line - in the Ribociclib group - 20, Palbociclib -22 (p=0.253). In the 3rd and later lines - the mPFS in the Ribocilib group - 19 (6.9-31.1), Palbociclib -14 (7.3-20.7) (p=0.775). The hematologic toxicity occurred significantly more often with Palbociclib (p=0.00001). As for Ribociclib, an undesirable event such as asthenia was prominent. The influence of dose-limiting toxicity on mPFS has been noted. In the presence of dose-limiting toxicity in the Palbociclib group, the mPFS has not been reached and is significantly different from the mPFS in patients who did not have a dose reduction. Conclusions: The analysis showed that Ribociclib is more often prescribed in the 1 line for patients with a higher 'tumor burden' and with primary hormone resistance. In relation to this, we observe a decrease in such an indicator as the frequency of overall response in this group of patients compared to Palbociclib.

Docetaxel plus cis-platinum chemotherapy for the posterior line treatment of metastatic castrate-resistant prostate cancer with elevated neuron specific enolase.

e17042 Background: This Study is to figure out the efficacy of docetaxel plus cis-platinum (DC) chemotherapy as the posterior line treatment for metastatic castrate-resistant prostate cancer (mCRPC) with elevated neuron specific enolase (NSE). Methods: We performed a retrospective study of men suffered mCRPC with elevated NSE from December 2019 to August 2022. All men were treated with DC after resistance to previous therapy. Biochemical progression-free survival (BPFS) were depicted using Kaplan-Meier curves. Results: 10 mCRPC patients aged 65.8 ± 7.0 (Median: 66.5, Range: 52-77) received DC for 1 to 10 cycles (Median: 7). Of the 10 patients, 2, 1 and 7 patients underwent DC as the 2nd, 3rd and 4th line therapy, respectively. 3 patients received previous chemotherapy and 9 patients underwent new endocrine therapy. All patients suffered bone metastases, 6 patients suffered distal lymph node metastases and 3 patients suffered visceral metastases. 6.1 ± 2.8 cycles (Median: 7, Range: 1-10) of DC were finished. Median follow-up time was 6.5 (Range: 2-11) months. After DC, Prostate specific antigen (PSA) response rate (&gt;50%) was 90% and NSE response rate (&gt;50%) was 60%. Median BPFS was 10 (Range: 4-18) months. Pain score declined from 6.6 ± 3.1 to 3.9 ± 2.4 evidently (P &lt; 0.05) and Karnofsky (KPS) score increased from 80 ± 14.9 to 91 ± 16.0 significantly (P &lt; 0.05). Of the 6 patients assessed by radiological examinations, 5 patients experienced partial response and 1patient achieved stable disease. Conclusions: Our pilot study showed docetaxel plus cis-platinum chemotherapy was effective for the mCRPC patients with elevated NSE resistant to multi-line therapies, indicating the potential value of DC for mCRPC with neuroendocrine differentiation.

Prevalence of recreational and medical cannabis products in Veterans and their interactions with cancer directed pharmacological treatment.

3089 Background: There is proliferation of literature about the medical uses of cannabis products in patients with malignancies. However, research is lacking as to how cannabis use is affecting the efficacy and toxicity of anti-neoplastic regimens It is known that cannabis is metabolized by the cytochrome p450 pathway, the same as many anti-neoplastic agents and thus there are concerns for potential interactions. Methods: We report an observational study to identify cannabis users in veterans receiving anti-neoplastic therapy in the Memphis VA health system, and to then analyzed for potential interactions between cannabis product and the antineoplastic agent/s they were receiving. Data was collected via voluntary surveys. We then reviewed their charts, recorded the anti- neoplastic medications they were receiving, and evaluated for potential interactions with cannabis, based on a literature review of potential interactions. Results: In this study, 132 veterans agreed to participate. 50 of them acknowledged recent use of cannabis products within the last 90 days. Thus, the prevalence of cannabis use amongst those who participated is 37.87%. Demographically, 10% of the cannabis users were female, and 90% were male, with the majority aged 60-65 (range 46-85). Predominantly, patients inhaled cannabis once daily (38%) as opposed to using a different formulation. The patients who admitted cannabis use were primarily on chemotherapy (42%), while 38% were on immunotherapy, 24% were on targeted therapy and 20% were on endocrine therapy. Some were on multiple drugs. Based on this data and a literature search for interactions, we found that a significant number of the patients in our Veteran’s oncology clinic have a potential increase in toxicity or decreased efficacy from their ant-neoplastic therapy due to CYP interactions with cannabis use. Conclusions: In this exploratory study, we discovered a notable prevalence of patients concurrently using cannabis products while on anti-neoplastic treatment. Given that many anti-neoplastic agents undergo metabolism via the CYP450 pathway, our data suggest that a moderate number of patients in our clinic risk less than optimal outcomes due to concurrent cannabis and anti-neoplastic therapy. To delve deeper, we intend to conduct a follow-up study, focusing on specific adverse effects, hospitalizations, and the subsequent need for 2nd and 3rd line treatments in patients using cannabis. A much larger multi-institutional study should be considered to provide statistically relevant data on true outcomes. [Table: see text]