Predictive Markers Of Response Research Articles

Dynamic monitoring of ctDNA to predict response to immunotherapy plus chemotherapy in BTC.

e14550 Background: The survival of advanced BTC patients, especially those who received palliative surgery was unfavorable and the treatment options for these patients were limited. Immune checkpoint inhibitors （ICIs） were approved for the treatment for patients with BTC. However, only a subset of patients benefits from this treatment and the clinical performance was not well evaluated. Therefore, there is a need for a method that can promptly and dynamically evaluate the efficacy of patients receiving immunotherapy. Previous studies have shown that ctDNA can detect relapse and monitor therapeutic efficacy in diverse cancer. We performed a retrospective observational study examining whether ctDNA could serve as a biomarker to track treatment responses in BTC. Methods: Patients diagnosed with advanced BTC and who received ICI were screened and enrolled between January 2020 and October 2022 at Shanghai Eastern Hepatobiliary Surgery Hospital. The key inclusion criteria for patients were as follows: a recent diagnosis of gallbladder/intrahepatic cholangiocarcinoma; received immunotherapy (or plus target therapy); received ctDNA NGS monitoring. Patients were excluded based on the following criteria: patients with multiple primary cancers, or had history of allogeneic organ transplantation or allogeneic hematopoietic stem cell transplantation, or HIV infected person, or severe infections with poor clinical control. Blood samples were collected for ctDNA analysis at the beginning, after 3 and 6 months of ICI treatment. Results: The median age was 66 (42-79), and 12 (44.4%) were male. The ECOG score was mostly 1 (23 cases, 85.2%). Most of the patients were stage IV (21, 77.8%), and there were 5 (18.5%) stage III patients, 1 (3.7%) stage II and 0 (0%) stage I patients. The BTC patients were mostly histologically diagnosed with gallbladder cancer (GBC, 11, 40.7%) and intrahepatic cholangiocarcinoma (8 cases, 29.63%). There were 55% patients received palliative surgery. Among all mutations detected in ctDNA, the mutation status of KRSA at baseline was detected to be related to patient survival, and the KRAS-mut group had a significantly lower probability of overall survival compared to the KRAS-WT group (p=0.014). Patients with more than 3 new mutations during treatment presented a significantly unfavorable OS (p=0.025) and a trend of unfavorable PFS (p=0.062). Besides, patients with increasing amounts of total ctDNA during dynamic monitoring showed significantly worse overall survival probability (p=0.023). Conclusions: Mutation of KRAS in tissues at baseline might be a predictive marker of response to immunotherapy and survival prognosis. ctDNA as a compose biomarker can monitor treatment responses and predict overall survival. Considering the small size and the relatively short follow-up time of our cohort, further prospective and randomized controlled studies are indispensable.

Use of tumor mutational burden as a predictive marker of response to immuno-oncology agents: Initial experience at an academic center in Venezuela.

e14613 Background: Tumor Genetic Profiling (TGP) is steadily becoming the mainstay of clinical practice, namely, the Tumor Mutational Burden (TMB) is believed to be a predictor of efficacy to treatment with Immune Checkpoint Inhibitors (ICI). Recent studies have demonstrated that a high-TMB score (TMB-H) is associated with higher response rates. In Venezuela, the evidence of mutational frequencies obtained by TGP is limited, herein we present our experience at a large educational institution. Methods: In this retrospective cohort, we included 80 subjects with different types of solid tumors to whom TGP was performed to determine TMB values, as well as other associated mutations, between January 2020 and December 2022. In a subgroup of subjects in which ICI therapy was formally indicated (n=21) we used the Objective Response Rate (ORR) as a primary endpoint, recorded as the percentage of patients who had a complete response (CR) or partial response (PR), based on the RECIST v1.1 criteria. Treatment options included Ipilimumab, Atezolizumab, Pembrolizumab and Nivolumab, as either monotherapy or in combination according to the corresponding NCCN guidelines. A TMB-H score was defined as greater than 10 mutations per megabase. Results: A TMB-H was found in 11 samples, subjects with lung cancer accounted for the majority of this finding (n=6). The most frequently reported mutations were BRCA 1-2 (n=9), KRAS (n=9), and EGFR (n=4). In the subgroup treated with ICI, the ORR of the group with TMB-H was 71.42%, while in those with low TMB it was 35.71%. The differences in response rates, with a 95% CI, yielded a p=0.122; which corresponds to a statistically non-significant outcome. Conclusions: This study constitutes the first country-level documentation of what is becoming a standardized practice, and contributes to the growing evidence regarding the use of TMB-H as a marker to predict response to treatment. Our small cohort demonstrated a high ORR to ICI when a TMB-H score is present, albeit to a non-statistically significant degree. We will continue to follow-up these patients, as well as to include newly sequenced samples for TGP, to give greater statistical power to these findings.

A phase 1 first-in-human clinical trial of HMBD-002, an IgG4 monoclonal antibody targeting VISTA, in advanced solid tumors.

TPS2664 Background: V-domain Ig Suppressor of T-cell Activation (VISTA), is an immune checkpoint regulator found on tumor, myeloid, and other immune cells. Its presence has been shown to enhance tumor growth, create an immunosuppressive microenvironment, and may potentially contribute to developing resistance to anti-CTLA-4 and anti-PD-1/PD-L1 therapies. Therefore, VISTA represents a promising therapeutic target. HMBD-002, a non-depleting, high-affinity IgG4 monoclonal antibody against VISTA, has demonstrated significant inhibition of tumor growth in preclinical studies, both as a monotherapy and in combination with pembrolizumab. HMBD-002 acts by increasing T-cell activity and reprogramming the suppressive tumor microenvironment to a proinflammatory antitumor phenotype. Cancer types including triple-negative breast cancer (TNBC) and non-small cell lung cancer (NSCLC), which exhibit high expression levels of VISTA in the TME, provide a rational basis for exploring these indications in clinical studies. Methods: This phase 1/2, open-label, multi-center trial is being conducted in two parts to evaluate the safety, pharmacokinetics, and antitumor activity of HMBD-002, an IgG4 monoclonal antibody targeting VISTA, as a monotherapy and in combination with pembrolizumab. The dose-escalation stage (Part 1) will follow a standard 3 + 3 study design with weekly dosing and adaptive dose-escalation increments to determine the recommended phase 2 dose (RP2D) for subsequent disease-directed studies. Monotherapy escalation cohorts 1-4 have been completed without dose-limiting toxicities (DLTs). The dose-expansion stage (Part 2) will evaluate the antitumor activity of HMBD-002 alone and combined with pembrolizumab in patients with TNBC and NSCLC, as well as a wide range of VISTA-expressing malignancies. Exploratory analyses will be conducted to assess treatment-induced immunological changes both systemically and within the tumor microenvironment. Longitudinal blood samples will be evaluated using single-cell mass cytometry (CYTOF) and multiplex cytokine analysis. Pre and on-treatment tumor tissue samples will be assessed for VISTA & PD-L1 expression using IHC. Additionally, tumor tissues will be evaluated using multiplex immunofluorescence, and gene-expression profiling. Biomarker correlations with clinical outcomes may lead to the identification of predictive markers of response to HMBD-002. Clinical trial information: NCT05082610 .

Initial results from the phase 2A trial of visugromab (CTL-002) + nivolumab in advanced/metastatic anti-PD1/-L1 relapsed/refractory solid tumors (The GDFATHER-TRIAL).

2501 Background: Increasing evidence has emerged that Growth and Differentiation Factor 15 (GDF-15) plays a critical role in tumoral immunosuppression. Apart from blocking immune-cell entry into the tumor microenvironment, GDF-15 also has a major impact on the formation of the immune synapse. Many tumors overexpress GDF-15 and have hijacked this mechanism to block immunotherapy success. Here we report initial results of the phase 2a part of the GDF-15 neutralizing antibody visugromab (CTL-002) in combination with nivolumab. Methods: At cut-off date (06-Feb-2023), a total of 79 pts with advanced/metastatic solid tumors have been enrolled and were treated with visugromab 10 mg/kg Q2W in combination with nivolumab 240 mg Q2W in the phase 2a expansion part of the CTL-002-001 phase 1/2 trial (NCT04725474). The primary endpoint of the phase 2a expansion is the Objective Response Rate (ORR) according to RECIST v1.1. Tumor assessments were performed every 8 weeks and patients had to be relapsed/refractory to a minimum of 12 weeks continuous prior anti-PD1/-PD-L1-containing therapy, with relapse or progression after initial response to have occurred on continued anti-PD1/PDL1 therapy. Phase 2a contained several indication-specific expansion cohorts (e.g. NSCLC, Bladder Cancer, HCC, Melanoma), following a Simon 2-stage design, and two cohorts were already expanded to stage 2 in light of initial responses observed. An additional expansion cohort is a pan-tumor basket cohort with sequential tumor biopsies for biomarker analyses. A key component of this multi-arm trial is an extensive translational research program to identify response-predictive tumor and immunologic marker and pattern. Results: Across several expansion cohorts, durable tumor regressions were observed, including several partial responses (PR) and one complete response (CR) in heavily pretreated patients with a median of 3 prior lines of systemic therapy for advanced/metastatic cancer, including at least one prior line of anti-PD1/-PD-L1 therapy and being relapsed/refractory to it. In NSCLC, so far 13 pts underwent at least one tumor assessment. Of these, two experienced a PR (one confirmed), three additional tumor shrinkage > 5% (TS) and one longer-term SD (LT-SD) were seen. In bladder cancer, 7 patients underwent at least one tumor assessment with 1 confirmed CR, 1 TS and 1 LT-SD. The treatment was overall well tolerated. Conclusions: These initial phase 2a results demonstrate promising clinical activity for the combination of visugromab and nivolumab in heavily pretreated anti-PD1/-PD-L1 refractory/relapsed solid tumor patients across various indications, including induction of confirmed partial and complete responses. Extensive biomarker evaluations are ongoing to further evaluate potential response-predictive markers identified during phase 1 development. Clinical trial information: NCT04725474 .

Combined expression of HIF1α, VEGF and HER2 predicts metastasis, relapse and response to combination chemotherapy in colorectal cancer patients

Addressing a single target is the frequent development of drug resistance followed by cancer relapse and treatment failure. Therefore, assessment of simultaneous expression of target molecules is essential to choose the optimal combination therapy for each colorectal cancer patient.This study aims to evaluate the immunohistochemical expression of HIF1α, HER2 and VEGF and to clarify their clinical significance as prognostic factors and predictive markers of FOLFOX (combination chemotherapy inclusive of Leucovorin calcium, Fluorouracil and Oxaliplatin response).Marker expression was retrospectively evaluated by immunohistochemistry in 111 patients with colorectal adenocarcinomas from south Tunisia, followed by statistical analysis.The immunohistochemical staining revealed that 45 %, 80.2 %, 86.5 % and 25.5 % of specimen were positive for nuclear, cytoplasmic HIF1α expression, VEGF and HER2 respectively. Nuclear HIF1α and VEGF were associated with worst prognosis while cytoplasmic HIF1α and HER2 were correlated with favourable prognosis. Multivariate analysis confirms the association between nuclear HIF1α, distant metastasis, relapse, FOLFOX response and 5 years overall survival. HIF1α positivity and HER2 negativity were significantly associated to short survival. Combined immunoprofiles HIF1α+/VEGF+, HIF1α+/HER2-, HIF1α+/VEGF+/HER2- were associated to distant metastasis, cancer relapse and short survival. Interestingly, our findings confirmed that patients bearing a HIF1α positive tumor were significantly more resistant to FOLFOX compared to negative ones (p = 0.002, p ≤ 0.001). Positive expression of HIF1α and VEGF, or decreased expression of HER2 was each associated with poor prognosis and short overall survival. In summary, we found that expression of nuclear HIF1α, alone or combined with VEGF and HER2 serves as a predictive marker of poor prognosis and FOLFOX response in colorectal cancer from south Tunisia.

High Expression of Tumor HLA-DR Predicts Better Prognosis and Response to Anti-PD-1 Therapy in Laryngeal Squamous Cell Carcinoma

BackgroundHLA-DR is expressed in epithelial and several types of tumor cells. However, the correlation between tumor-expressed HLA-DR (teHLA-DR) and patient outcome as well as its regulation on the tumor microenvironment (TME) of laryngeal squamous cell carcinoma (LSCC) are yet to be elucidated. MethodsHematoxylin and eosin (HE) staining were performed to define the tumor nest and stroma of LSCC tissue microarrays. teHLA-DR tumor cell, CD4+ and CD8+ tumor-infiltrating T lymphocytes (TITLs) were obtained and analyzed through double-labeling immunofluorescence and immunohistochemical staining. The recurrence-free (RFS) and overall survival (OS) curves were plotted using the Kaplan-Meier method and tested by the log-rank test method. Expression of teHLA-DR+ tumor cells and infiltration of T lymphocytes and their corresponding subgroups were analyzed by flow cytometry using fresh LSCC tissue samples. ResultsOur research discovered elevated expressions of multiple MHC-II-related genes in tumor compared to the adjacent normal tissue samples of LSCC patients. We also found that patients in the teHLA-DR high-expression group (teHLA-DRhigh) tend to have less tumor recurrence and better survival outcomes compared to those in the teHLA-DRlow group. Intriguingly, teHLA-DR+ tumor cells had significantly higher PD-L1 and PD-L2 expression and their TME showed increased infiltrated T lymphocytes (TITLs). Flow cytometry analysis and IHC staining indicated that CD4+ TITLs but not CD3+ total TITLs or CD8+ TITLs were significantly enriched in teHLA-DR+ tumors. ConclusionsteHLA-DR may be a predictive marker for favorable prognosis and response to anti-PD-1/PD-L1 therapy of LSCC, possibly due to the increased CD4+ TITLs in the TME.

Predictive markers of response to immunotherapy in uterine cancer

Рак тела матки (РТМ) является одним из ведущих причин онкологической заболеваемости у женского населения. При этом, до четверти всех случаев РТМ представлены распространенным процессом, при котором до недавнего времени отсутствовали эффективные подходы к лечению после прогрессирования заболевания на стандартной химиотерапии. Ингибиторы контрольных точек (ИКТ) позволили значительно расширить терапевтические опции для этой группы пациентов. Несмотря на то, что терапия ИКТ при РТМ продемонстрировала многообещающие результаты, данное заболевание является гетерогенным, что обуславливает важность изучения потенциальных предикторов эффективности лечения. В обзорной статье представлены данные о молекулярных основах ключевых предиктивных маркеров ответа на иммунотерапию при РТМ и их клиническом значении при рецидивирующем и метастатическом заболевании.

Expression of LAG-3 on B-lymphocytes as a marker for prediction of response to therapy in patients with chronic lymphocytic leukemia

Purpose: to study the level of LAG-3 expression on B-lymphocytes and the feasibility of using it as a marker for predicting response to therapy in patients with chronic lymphocytic leukemia (CLL).Material and Methods. The study included 40 patients with newly diagnosed CLL. All patients were divided into two groups: group I: patients with Binet stage A, who did not receive therapy and group II: patients with Binet stage C, who received immunochemotherapy in RB and FCR regimes. According to the treatment regimen and hematological response to therapy, 4 subgroups were distinguished: IIA-RB, IIA-FCR, IIB-RB, and IIB-FCR. The control group consisted of 20 people matched in age and gender without cancer. The immunophenotype, level of B-lymphocytes, LAG-3 expression, and the minimal residual disease in group II after the 6th course of immunochemotherapy were initially determined in all groups by flow cytometry. The data were evaluated using Statistica 13.0.Results. Compared to the control group, the LAG-3 expression on B-lymphocytes was found in all groups of CLL patients before treatment. The expression level was higher in patients with Binet stage C than in patients with Binet stage. The data demonstrated differences in the level of LAG-3 expression in patients with different hematological responses to therapy. The initially higher level of LAG-3 expression on B-lymphocytes was observed in patients with Binet stage C CLL with an unfavorable response to therapy. A good hematological response was found can be achieved at the level of LAG-3 expression within 14.57 ± 0.66 % regardless of the therapy regimen, and unfavorable response to therapy at the level of 41.95 ± 1.62 %.Conclusion. The initial level of LAG-3 expression on B-lymphocytes in patients with CLL can be used as a marker for predicting and monitoring response to treatment, regardless of the immunochemotherapy regimen used.

Response to primary chemoradiotherapy of locally advanced oropharyngeal carcinoma is determined by the degree of cytotoxic T cell infiltration within tumor cell aggregates.

Effective anti-tumor immune responses are mediated by T cells and require organized, spatially coordinated interactions within the tumor microenvironment (TME). Understanding coordinated T-cell-behavior and deciphering mechanisms of radiotherapy resistance mediated by tumor stem cells will advance risk stratification of oropharyngeal cancer (OPSCC) patients treated with primary chemoradiotherapy (RCTx). To determine the role of CD8 T cells (CTL) and tumor stem cells for response to RCTx, we employed multiplex immunofluorescence stains on pre-treatment biopsy specimens from 86 advanced OPSCC patients and correlated these quantitative data with clinical parameters. Multiplex stains were analyzed at the single-cell level using QuPath and spatial coordination of immune cells within the TME was explored using the R-package Spatstat. Our observations demonstrate that a strong CTL-infiltration into the epithelial tumor compartment (HR for overall survival, OS: 0.35; p<0.001) and the expression of PD-L1 on CTL (HR: 0.36; p<0.001) were both associated with a significantly better response and survival upon RCTx. As expected, p16 expression was a strong predictor of improved OS (HR: 0.38; p=0.002) and correlated with overall CTL infiltration (r: 0.358, p<0.001). By contrast, tumor cell proliferative activity, expression of the tumor stem cell marker CD271 and overall CTL infiltration, regardless of the affected compartment, were not associated with response or survival. In this study, we could demonstrate the clinical relevance of the spatial organization and the phenotype of CD8 T cells within the TME. In particular, we found that the infiltration of CD8 T cells specifically into the tumor cell compartment was an independent predictive marker for response to chemoradiotherapy, which was strongly associated with p16 expression. Meanwhile, tumor cell proliferation and the expression of stem cell markers showed no independent prognostic effect for patients with primary RCTx and thus requires further study.

Energy Metabolism Is Altered in Radioresistant Rectal Cancer.

Resistance to neoadjuvant chemoradiation therapy is a significant clinical challenge in the management of rectal cancer. There is an unmet need to identify the underlying mechanisms of treatment resistance to enable the development of biomarkers predictive of response and novel treatment strategies to improve therapeutic response. In this study, an in vitro model of inherently radioresistant rectal cancer was identified and characterized to identify mechanisms underlying radioresistance in rectal cancer. Transcriptomic and functional analysis demonstrated significant alterations in multiple molecular pathways, including the cell cycle, DNA repair efficiency and upregulation of oxidative phosphorylation-related genes in radioresistant SW837 rectal cancer cells. Real-time metabolic profiling demonstrated decreased reliance on glycolysis and enhanced mitochondrial spare respiratory capacity in radioresistant SW837 cells when compared to radiosensitive HCT116 cells. Metabolomic profiling of pre-treatment serum samples from rectal cancer patients (n = 52) identified 16 metabolites significantly associated with subsequent pathological response to neoadjuvant chemoradiation therapy. Thirteen of these metabolites were also significantly associated with overall survival. This study demonstrates, for the first time, a role for metabolic reprograming in the radioresistance of rectal cancer in vitro and highlights a potential role for altered metabolites as novel circulating predictive markers of treatment response in rectal cancer patients.

Abstract 954: PIK3CA mutation induces immunogenicity and increases the immune checkpoint inhibitor response in urothelial carcinoma

Abstract Background: Immune checkpoint inhibitor (ICI) has been widely used for advanced urothelial carcinoma (mUC) globally. However, little is known about the potential predictive factors of the ICI response. The study aims to identify a potential biomarker to predict ICI response in UC. Material and methods: We performed targeted next generation sequencing (NGS) of pre-treatment tumor Formalin-fixed paraffin-embedded (FFPE) samples by using ACTOnco® comprehensive genomic panel (ACT Genomics, Co. Ltd., Taiwan) to target all coding exons of 440 cancer-associated genes. We retrospective analyzed a total of 70 patients with mUC underwent at least one cycle of ICI with or without chemotherapy at Kaohsiung Chang Gung Memorial Hospital between April 2016 and November 2019. Patients were defined as responders (CR/PR) and non-responders (SD/PD) according to the RECIST evaluation. TMB was estimated by calculating all coding exons variants, including somatic non-synonymous, frameshift and splice site variants. Results: A total of 55 UC tumor samples passed QC process were analyzed. The median age was 67, and 74% of all patients were in ICI monotherapy cohort. The most common genetic alterations were TP53 (66%), KMT2D (42%) and MUC16 (32%). In ICI monotherapy group, responders had a higher TMB than non-responders (median 19.5 vs 6.5 mt/Mb; p = 0.004). Patients with high TMB on ICI monotherapy had a significantly longer PFS (14.6 vs 2.9 months; p = 0.003) and OS (NR vs 11.0 months; p = 0.01) than TMB low patients. We found that PIK3CA (32% vs 3%), UBR5 (23% vs 0%), PRKDC (32% vs 10%) were the top 3 enriched mutations between responders and non-responders. All PIK3CA mutation were hot-spot mutation, and tumor harbored PIK3CA mutation had a median high TMB than PIK3CA wild type in UC, NSCLC, breast and colorectal cancer. By analyzing tumor and immune microenvironment by TISIDB platform, UC with PIK3CA mutation was highly associated with increasing MHC molecules expression (HLA-A, HLA-B, HLA-C, B2M) in UC and gastric cancer. In vitro study, we confirmed anti-PDL1 with PBMC had a synergistic cytotoxicity in PIK3CA mutated cell lines (BFTC-909, AGS and DLD-1), but not in PIK3CA wild type cell lines (T24, RT4, UMUC-14, N87 and SW837). Knocking-down PIK3CA expression diminished the cytotoxic activity of anti-PDL1 with PBMC in BFTC-909 and AGS cells. PIK3CA down-regulation inhibits antigen presentation machinery process. We confirmed that knockout PIK3CA gene function by CRISPR/Cas9 decreased MHC1 and B2M molecules expression, and showing modest resistance to ICI than mice with preserved PIK3CA function. Conclusion: UC with PIK3CA missense mutation exhibits a higher immunogenicity through enhancing MHC class I and B2M expression. PIK3CA missense mutation is a potential predictive marker for ICI response in UC. Citation Format: Harvey Yu-Li Su, Ling-Yi Xiao, Shih-Yu Huang, Li-Chung Chang, Yi-Hua Chen, Hao-Lun Luo, Chia-Ling Wu. PIK3CA mutation induces immunogenicity and increases the immune checkpoint inhibitor response in urothelial carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 954.

Abstract 1848: A novel murine tumor model for assessing the efficacy of neoadjuvant dosing regimens for immunotherapies

Abstract Immunotherapy extends survival for patients with several cancer types, especially in combination with surgery. Consistently, recent evidence demonstrated improved efficacy with neoadjuvant/adjuvant pembrolizumab in combination with debulking surgery in patients with melanoma (Patel et al, ESMO 2022). However, evaluating these regimens preclinically is challenging due to the rarity of appropriate models. To address this question, we have established a novel murine tumor model involving partial tumor-debulking surgery which enables monitoring of the temporal response to immunotherapy in neoadjuvant, neoadjuvant/adjuvant and adjuvant settings. We utilized the MC38 tumor model which represents a ‘hot’ tumor microenvironment (TME), analogous to the TME in melanoma patients. MC38 tumor-bearing C57Bl/6 mice were subcutaneously administered murine nemvaleukin alfa (mNemva), an engineered IL-2 immunotherapy that selectively activates and expands cytotoxic NK cells and antigen-experienced CD8+ T cells. Partial tumor-debulking surgery was performed under general anesthesia 3 days after the first dose of mNemva along with neoadjuvant and neoadjuvant/adjuvant regimens, or 1 day before the first dose of mNemva in the adjuvant regimen. Whereas a single dose of mNemva was administered to mice in the neoadjuvant setting, four and five doses were given in the adjuvant and neoadjuvant/adjuvant settings, respectively. Two types of control groups were also evaluated: vehicle in combination with partial tumor-debulking surgery, and five doses of mNemva without the debulking procedure.The neoadjuvant/adjuvant treatment regimen was superior to neoadjuvant or adjuvant in all aspects with 28% of animals exhibiting complete responses, tumor growth inhibition (TGI) of 82% and prolonged median survival from 17 to 27 days. While the neoadjuvant regimen yielded 10% complete responders, TGI was only 7% and median survival was not extended. No complete responders were observed with the adjuvant regimen. Importantly, tumors in both control groups grew to the predefined endpoint irrespective of treatment and surgical procedure. Following tumor re-challenge, 75% of prior complete responders were protected, suggesting that mNemva induced immunological memory in most mice. Moreover, long-term immunity was 100% effective if the re-challenge experiment was performed out to 5 months after the last dose of mNemva had been administered.Therefore, the neoadjuvant/adjuvant dosing regimen exhibits a superior response in this preclinical model, when combined with partial tumor-debulking surgery, similar to the recent findings in patients with melanoma that were treated with pembrolizumab. In addition, this strategy enables sampling of tumor tissue for analyses of the immune response and the identification of potential predictive markers of response. Citation Format: Sara Prijic, Jason Wood, Arthur Liberzon, Harikiran Nistala, Raymond J. Winquist, Jared E. Lopes. A novel murine tumor model for assessing the efficacy of neoadjuvant dosing regimens for immunotherapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1848.

Abstract 6113: Predictive value of HRD testing on the response to PARP inhibitors in ovarian cancer preclinical models

Abstract Since the introduction of Poly(ADP-ribose)polymerase inhibitors (PARPi) in the clinical practice, the search of biomarkers to distinguish homologous recombination defective tumors, eligible for PARPi therapy, has become a priority. Along with testing for BRCA mutations, commercially available tests, such as MyChoice® assay (Myriad Genetics), exploit scores calculated from genomic scarring patterns. Platinum sensitivity is also considered a surrogate marker of PARPi efficacy. However, the prognostic accuracy remains a challenge because these tests can only infer homologous recombination functionality but not directly measure it. Moreover, the association between platinum and PARPi response is imperfect since not all platinum-sensitive patients benefit from PARPi and PARPi response has been observed in platinum-resistant cancers. A panel of 20 high-grade serous ovarian carcinoma patient-derived xenografts (OC-PDXs) were molecularly and pharmacologically characterized to assess predictive markers of response to PARPi therapy. Genomic DNA and total RNA were sequenced searching for BRCA1/2 alterations or sent to Myriad Genetics to obtain HRD scores. Absolute BRCA1 mRNA expression was measured by RT-qPCR. OC-PDXs growing subcutaneously or orthotopically in nude mice were treated with the PARPi olaparib or platinum to test drug sensitivity. The molecular analyses revealed that nearly 90% (17/20) of OC-PDXs are TP53 mutated, thus reflecting the clinical scenario. 50% (10/20) of OC-PDXs harbor mutations in BRCA1/2 genes, while 10% (2/20) lack BRCA1 mRNA expression. MyChoice® assay showed that 65% (13/20) of OC-PDXs score higher than 42, which is the currently used threshold to classify HRD-positive tumors. Correlation analysis confirmed that the HRD score is highly associated with the presence of mutations in BRCA1/2 genes. Preclinical testing showed that 50% of OC-PDXs are sensitive to olaparib. Notably, in 20% of cases the therapeutic response was incorrectly predicted considering either BRCA1/2 mutations or HRD score. 30% of platinum-sensitive OC-PDXs were not responsive to olaparib, confirming that platinum sensitivity not always correlate with response to PARPi. Our results show that tumor genotype and HRD score are strongly associated. Further analyses are required to improve the predictive value of currently used tests to select patients who will benefit the most from PARPi treatment while sparing toxicities in patients that are unlikely to respond. Citation Format: Giulia Dellavedova, Alessandra Decio, Laura Formenti, Luca Porcu, Carmen Ghilardi, Raffaella Giavazzi, Maria Rosa Bani. Predictive value of HRD testing on the response to PARP inhibitors in ovarian cancer preclinical models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6113.

Abstract 2129: Predictive response biomarkers from Phase I clinical trial of a SIRPalpha inhibitor BI765063, stand-alone and in combination with ezabenlimab, a PD1 inhibitor, in patients with advanced solid tumors

Abstract Background: Signal Regulatory Protein α [SIRPα] is an inhibitory membrane receptor expressed by myeloid cells (macrophages and myeloid-derived suppressor cells, MDSCs) and specifically binds to CD47. BI765063 is a selective anti-SIRPα monoclonal antibody acting as a checkpoint inhibitor of SIRPα/CD47 axis, promoting anti-tumor immunity through increase of dendritic cells antigen presentation, enabling MDSC differentiation, potentiating macrophage phagocytic/inflammatory properties and reinstating myeloid cell chemokine secretion and human T cell migration1. The escalation phase I study in advanced solid tumor patients (pts) showed preliminary efficacy as monotherapy (MONO), and in combination with PD-1 inhibitor ezabenlimab (COMBO). Our goal was to investigate BI765063 predictive response markers. Methods: A total of 68 patients (50 in MONO arm, 18 in COMBO arm) have been enrolled. Receptor occupancy (RO) was determined on peripheral CD14+ monocytes. Tumor biopsies were collected before treatment from 44 pts (62%) representative of overall population. Tumor microenvironment (TME) was analysed using a Brightplex® IHC panel comprised of CD68+ macrophages, CD11b+myeloid cells, SIRPα, and CD47. NanoString tumor profiling used PanCancer IO360 panel. Results: One partial response (PR) in MONO (Hepatocellular carcinoma HCC), and 4 PRs in COMBO (2 endometrial cancer, 1 HCC, 1 iRecist PR in MSS CRC) were observed2. BI 765063 full RO saturation was achieved in V1/V1 pts treated with doses of 6 mg/kg and higher, while V1/V2 patients showed a more heterogeneous RO ranging between 40-80%, reaching an apparent saturation at doses of 12 mg/kg and higher. Comparison of pts by best response showed that baseline tumor SIRPα+ expression in CD68+ macrophages and in CD11b+ myeloid cells were higher in PR versus PD pts. High percentage of CD11b+SIRPα+ myeloid cells at baseline significantly correlated with better OS (p=0.023), while CD68+SIRPα+ macrophages high expression in TME showed a trend for a better OS but did not reach statistical significance. The CD47 tumor expression did not correlate with the OS. Responder signature based on the top deregulated genes in responders versus non-responders at baseline was devised to sort relevant TCGA cohorts and showed strong positive correlation with TME MDSC infiltration (p≤0.0001). Conclusions: High levels of CD11b+SIRPα+ myeloid cells in TME at baseline, but not CD47 tumor expression, correlates with longer survival while MDSC signature in TME at baseline correlates with clinical response. Thus, suggesting that MDSCs expressing SIRPα in TME could represent a predictive efficacy biomarker.

Development of an artificial intelligence-derived histologic signature associated with adjuvant gemcitabine treatment outcomes in pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) has been left behind in the evolution of personalized medicine. Predictive markers of response to therapy are lacking in PDAC despite various histological and transcriptional classification schemes. We report an artificial intelligence (AI) approach to histologic feature examination that extracts a signature predictive of disease-specific survival (DSS) in patients with PDAC receiving adjuvant gemcitabine. We demonstrate that this AI-generated histologic signature is associated with outcomes following adjuvant gemcitabine, while three previously developed transcriptomic classification systems are not (n= 47). We externally validate this signature in an independent cohort of patients treated with adjuvant gemcitabine (n= 46). Finally, we demonstrate that the signature does not stratify survival outcomes in a third cohort of untreated patients (n= 161), suggesting that the signature is specifically predictive of treatment-related outcomes but is not generally prognostic. This imaging analysis pipeline has promise in the development of actionable markers in other clinical settings where few biomarkers currently exist.

Neutrophil-to-lymphocyte ratio as an early marker of outcomes in patients with recurrent oral squamous cell carcinoma treated with nivolumab

The immune checkpoint inhibitor (ICI), nivolumab, has revolutionised the treatment of recurrent and metastatic oral cancer. However, the response rate to ICIs remains low, and identifying predictors of nivolumab response is critical. Although the neutrophil-to-lymphocyte ratio (NLR) has been suggested as a predictive marker of nivolumab response in patients with various types of cancer, its utility in oral squamous cell carcinoma (OSCC) has not been elucidated. In this retrospective multicentre cohort study, we evaluated the association between NLR and outcome of nivolumab treatment in 64 patients with OSCC treated between 2017 and 2020. The objective response and disease control rates were 25.1% and 32.9%, respectively. The rates for complete and partial responses were 15.7% (10/64) and 9.4% (6/64), respectively; stable and progressive disease rates were 7.8% (5/64) and 67.1% (43/64), respectively. Complete and partial responses were classified as responders, and stable and progressive diseases were classified as non-responders. The median (range) pre-treatment NLR among responders was 4.3 (2.8–8.0), which decreased to 4.0 (2.6–6.3) after nivolumab treatment, and the median (range) pre-treatment NLR among non-responders was 5.1 (2.7–7.9), which increased to 6.4 (4.0–14.0) with tumour growth. Moreover, overall survival was significantly worse in the group with a higher post-treatment NLR (≥5) than in the group with a lower NLR (<5). Patients with a post-treatment NLR of ≥6 had worse outcomes for salvage chemotherapy following nivolumab treatment. Thus, post-treatment NLR could be a useful marker for predicting the response to nivolumab treatment or salvage chemotherapy in patients with OSCC.

Circulating Tumor Cells Predict Response of Neoadjuvant Chemotherapy in Patients with Bladder Cancer: A Preliminary Study.

This study aimed to explore the existence of circulating tumor cells (CTCs) in patients with muscle-invasive bladder cancer (MIBC) and their predictive potential for response to neoadjuvant chemotherapy (NAC). From 33 blood samples of MIBC patients, CTCs were isolated by cell surface markers and enriched by the IsoFlux™ device, followed by morphological and immunofluorescent identification. CTCs were detected at baseline in all samples. Immunofluorescence confirmed the tumor origin. MIBC patients were stratified by NAC response into the disease control (DC) and progressive disease (PD) groups. In the DC group, the number of CTCs decreased significantly after four courses of NAC (p < 0.0001). CTC counts in 7.5 mL after four NAC cycles were highly correlated with postoperative pathological T stage (p < 0.0001). Our study demonstrated that CTCs might represent a valuable predictive marker for NAC response in MIBC. CTC detection in MIBC patients could allow early arrangement of radical cystectomy for NAC non-responders to prevent disease progression while receiving the NAC courses.

Abstract P4-06-08: HER2-protein expression is a predictive marker for treatment response in patients with HER2-positive breast cancer who received neoadjuvant chemotherapy with dual HER2-blockade

Abstract Background In HER2-positive breast cancer, pathologic complete response (pCR) has been known to be a surrogate marker for favorable prognosis after neoadjuvant chemotherapy. We aimed to identify the clinico-pathologic factors related to pCR in patients with HER2-positive breast cancer treated with neoadjuvant dual HER2-targeted therapy. Methods Two-hundred ninty-five patients with HER2-positive breast cancer who received neoadjuvant chemotherapy with docetaxel, carboplatin, trastuzumab, and pertuzumab (TCHP) were included in this retrospective analysis. We assessed the association between age, clinical T, N stage, pathologic factors such as histologic grade, hormone receptor (HR), tumor-infiltrating lymphocytes, the membranous expression of HER2 protein evaluated by immunohistochemistry (IHC) and pCR. The fluorescent in situ hybridization (FISH) was performed in a tumor with HER2 score of 2+. Results Of the 295 patients, 195 (66.1%) achieved pCR (ypT0/is and ypN0). Besides, 240 (81.4%) patients were HER2 score of 3+, and 55 (18.6%) patients were HER2 score of 2+. The pCR was frequently observed in patients with HER2 score of 3+ (64 of 176 [73.3%]) than in those with HER2 score of 2+ (19 of 55 [34.5%]), regardless of HR status. After adjusting other clinicopathologic factors, the high HER2 protein expression was only an independent factor for pCR (adjusted OR 3.85, 95% CI, 1.66-8.91, p=0.002). Conclusions Our results suggest that high expression of HER2 protein assessed by IHC is important in predicting neoadjuvant TCHP response in HER2-positive breast cancer. Citation Format: Soong June Bae, Ji Soo Jang, Yoonwon Kook, Seung Ho Baek, Jee Hung Kim, Sung Gwe Ahn, Joon Jeong. HER2-protein expression is a predictive marker for treatment response in patients with HER2-positive breast cancer who received neoadjuvant chemotherapy with dual HER2-blockade [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-06-08.

Abstract P3-04-11: Longitudinal Changes of Contralateral Breast Mammographic Artificial Intelligence Algorithms Score in Ductal Carcinoma In Situ patient with Tamoxifen

Abstract Background Mammographic Artificial intelligence (AI) algorithms (Lunit insight MMG) draws attention as a diagnostic support tool for breast cancer detection. Lunit insight MMG provides a location suspected of breast cancer with a heatmap and a score reflecting the probability of the presence of suspicious areas. We investigated whether the Lunit insight MMG score is relevant for predicting the response to adjuvant tamoxifen. Methods Patients diagnosed with DCIS and underwent treatment at Seoul National University Hospital in 2010 were retrospectively enrolled. Clinical characteristics, tamoxifen use, survival data, and mammography images were extracted from the electronic medical records, and Lunit insight MMG scores were calculated retrospectively. We classified two groups according to tamoxifen treatment and compared the score change of contralateral breast from baseline to 5 years after surgery for DCIS. Change categories of Lunit insight MMG score included maintaining high risk, maintaining low risk, increasing from low risk to high risk, and decreasing from high risk to low risk. Results Of 100 patients, 50 (50%) had undergone tamoxifen treatment (group 1) and 50 (50%) had not (group 2). The median age of the patients was 48.4 years for group 1 and 51.5 years for group 2 (p=0.172). The median follow-up duration was 8.7 years for the whole cohort. Using Lunit insight MMG score, more patients in group 1 decreased in contralateral breast cancer (CBC) risk compared with group 2 (6.0% vs. 2.0%; P=.008). No patients in group 1 had an increase in CBC risk while 9 patients increased in group 2 (0% vs. 18.0%; P =.008). There was no ipsilateral breast cancer recurrence for the whole cohort, and two patients experienced contralateral invasive breast cancer in group 2. In two patients with CBC, the Lunit insight MMG score increased five years after surgery, one year and three years before the CBC diagnosis. Conclusions Longitudinal Changes of Mammographic AI algorithms Score may be a predictive surrogate marker for response to tamoxifen therapy in hormone receptor-positive DCIS. Citation Format: Changjin Lim, Eun Kyung Park, Hong-Kyu Kim, Eunhye Kang, Ji-Jung Jung, Hyunsu Yeoh, Jang-il Kim, Jung Whan Chun, Han-Byoel Lee, Hyeong-Gon Moon, Wonshik Han. Longitudinal Changes of Contralateral Breast Mammographic Artificial Intelligence Algorithms Score in Ductal Carcinoma In Situ patient with Tamoxifen [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-04-11.

Predictive models assessing the response to ustekinumab highlight the value of therapeutic drug monitoring in Crohn's disease.

Despite the therapeutic efficacy of Ustekinumab (UST) in Crohn's disease (CD), loss of response (LOR) is observed over time. This study aims to evaluate the impact of the UST pharmacokinetics (PK) at induction on clinical and endoscopic outcomes, as well as to find predictive markers of UST response. This retrospective study included 80 CD patients. Pharmacokinetics data (trough levels (TLs)) combined with clinical and biological parameters were fed into tailored logistic regression and tree-based ensemble techniques to predict clinical and endoscopic outcomes at one year of follow-up. TLs at week 16 were significantly lower among patients with moderate to severe endoscopic activity during the follow-up (p=0.04). The best model to predict endoscopic outcome was obtained at week 16 by Random Forest with an area under the receiver operating characteristic curve of 0.92±0.08, sensitivity 91% and specificity 75%, with key inputs such as lymphocyte and monocyte counts at week 8, and UST TLs and CRP at week 16. This real-world study confirms the relationship between early UST TLs and both clinical and endoscopic outcomes. Models were developed for the task of predicting clinical and endoscopic remission in CD patients treated with UST, highlighting the clinical relevance of UST TLs at week 16.