Background

The multipotent cancer stem-cell differentiates into tumor cell and mesenchymal stromal cell. Depending on the favorability of the tumor microenvironment and the metabolic demand, the ratio of stromal cells to tumor cells differ in the tumor microenvironment. Our study corelates the stromal content in molecular CAF whole body imaging with stroma to tumor ratio and outcomes of locally advanced esophageal cancer patients subjected to different radiation doses with platinum-based doublet chemotherapy regimens.

Methods

54 patients with locally advanced carcinoma of esophagus fitting into definitive chemoradiation therapy protocol were prospectively included in the study after institutional ethical clearance. All patients underwent molecular CAF whole-body imaging prior to initiation of definitive chemoradiation. The SUV Maximum at the primary site were corelated with the stromal content in histopathology. A stromal content of above 50 % was quantified as high stroma. SUV uptake of more than 7 was taken as High SUV. Patients received either 50Gy or 60Gy concurrent with platinum-based doublet chemotherapy. All patients were followed up for a period of minimum 2 years and the response assessment CAF imaging done at the end of 3 months from chemoradiation. The imaging-based response rates, disease free survival and overall survival at the end of 2 years were analyzed.

Results

The local and systemic control of patients with low stromal content is better than that with high stromal content. Amongst the high stromal content, Patients who received 60Gy had better local control rates than the ones that received 50Gy. In the low stromal content irrespective of radiation dose the local control was achieved. The systemic control and overall survival of high stromal content patients appears poorer necessitating the need for further studies on adjuvant, maintenance, or metronomic chemotherapy.

Table: 256MO

Conclusions

CAF Radiological imaging and pathological stromal content quantification could be a Bio-predictive independent tool for radiation dose escalation to augur, telltale responses and survival.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Lung cancer diagnostics has witnessed a paradigm shift owing to rapid developments of agents for targeted molecular therapy. However, in a resource constrained country like India, panel based NGS cannot be made available to the population at large. Integration of artificial intelligence in cancer is showing promise in early detection, prognostication and therapeutic outcomes. After optimal successes in breast and esophageal cancers, lung cancer diagnostics, poses an unmet need for the same. NGS requires tedious workflows causing inadvertent delays in final report. Integrating a fairly accurate AI to predict oncogene addiction may help clinicians triage patients for NGS.

Methods

Whole slide images of hematoxylin and eosin (H&E)-stained slides of lung adenocarcinoma were used and split into multiple tiles and annotated for acinar, lepidic, micropapillary, papillary, or solid patterns. The tiles were divided into training, validation, and testing subsets. Training was conducted using VGG16 network (Keras Applications). The trained model was tested/validated with two independent datasets consisting of 5946 tiles each. In the next stage, the demographic and genomic data were combined with patterns to predict the mutational status of oncogenes such as EGFR, ALK, and ROS1. Multiclass ML algorithms such as random forest, decision tree classifier, extreme gradient boosting (XGBoost), GridSearchCV, and RandomizedSearchCV were implemented.

Results

3420 cases were used for training and 100 for prospective validation. The final model trained on VGG16 deep learning network (Keras Applications) performed at 87.22% validation and 80.49% testing accuracy conducted on two separate datasets(5946 image tiles each). Multiclass ML algorithms exhibited an accuracy of more than 90% for the mutation prediction (EGFR: 90.6%, ALK: 85.2%, ROS:79.3%). The number of cases in ALK and ROS1 positive groups were few, leading to lesser accuracy.

Conclusions

This initial experience suggests that integrating AI in lung cancer genomics is a value addition in the toolkit of the oncologist, aiding in prompt institution of appropriate therapy as well as triaging patients requiring upfront chemotherapy vs. panel based NGS testing.

Legal entity responsible for the study

Ullas Batra.

Funding

Conquer Cancer Foundation.

Disclosure

All authors have declared no conflicts of interest.

Background

Immune checkpoint inhibitors (ICI), such as anti-PD-1, are used to facilitate anti-cancer function of effector T cells, which gives a passable treatment effect to patients, and CD8⁺ T cell exhaustion is a key impediment to ICI treatment. To improve anti-cancer effect of ICI, combination therapy is a potential strategy. Chemo-immunotherapy combination augments therapeutic efficacy of ICI, but elevated adverse effects are accompanied with patients. Thus, a tumor-microenvironment-on-chip (TMoC) system was developed to identify synergism with ICI therapy to maximize treatment efficacies while reducing the side effects.

Methods

To accurately imitate in vivo drug response, the tumor environment needs to be effectively preserved in vitro. Oxygen gradient is a critical mechanism that leads to cellular heterogeneity, resulting in T cell exhaustion and immunotherapy failure. Herein, to study the interaction between drugs and tumor-microenvironment, TMoC was designed with characteristics of molecular gradients, including oxygen, nutrients, and drugs. Besides, CD8⁺ T cells exhibited diverse behavior in different areas in TMoC, which was highly similar to in vivo conditions. Further, the efficacy of CD8⁺ T cells against tumors could also be quantified to assess the synergism of immune-chemotherapy combinations.

Results

TMoC had been shown to have an oxygen gradient. In the region of relative hypoxia, exhaustion of CD8⁺ T cells, anti-apoptosis of cancer cells, and drug resistance genes were significantly up-regulated. To improve the effects of anti-PD-1, we introduced several potential chemotherapy agents. A TGF-β1 receptor antagonist, and a MAPK pathway inhibitor were identified to synergize with anti-PD-1 in hypoxia or normoxia, respectively. Further, the drug evaluation in TMoC was confirmed to be the same in orthotopic preclinical cancer model.

Conclusions

Characteristics of in vivo tumor environment were perfectly preserved in TMoC, and responses to the drug combinations were highly similar. Thus, we concluded that TMoC could serve as a powerful tool to improve drug development, and clinical diagnosis purposes.

Legal entity responsible for the study

The authors.

Funding

Ministry of Science and Technology, R.O.C.

Disclosure

All authors have declared no conflicts of interest.

Background

Accumulating evidence points to the clinical relevance of the gut microbiota on outcome of immune checkpoint inhibitors (ICI). Recent work identified a gut oncomicrobiome signature centered by Akkermansia muciniphila (Akk) associated with favorable outcomes to ICI in advanced lung (NSCLC) and kidney (RCC) cancer patients.

Methods

ONCOBIOTICS (NCT04567446) provided shotgun metagenomics sequencing (MGS) of fecal samples from NSCLC and RCC during ICI in France and Canada. We reconstructed topological pearson networks within the microbial ecosystem of patients with overall survival >12 (responders: R) and <12 months (non-responders: NR). Networks of species interacting groups (SIG), notably two highly enriched in harmful (SIG1) or beneficial (SIG2) bacteria, were identified. Then, we computed TOPOSCORE, a monodimensional score based on SIG1/SIG2 ratio combined with Akk relative abundance. Multivariate analysis (MVA) was used to adjust for established prognostic factors.

Results

In two different cohorts with a total of 393 NSCLC, we could classify patients in R or NR using TOPOSCORE. The sensitivity, specificity, positive and negative predictive value were 80%, 47%, 67% and 63%, respectively. The MVA revealed that the TOPOSCORE was an independent prognostic factor (Table). In new cohorts of NSCLC (n=50) and RCC (n=83), TOPOSCORE outperformed PD-L1 and the International Metastatic RCC Database Consortium (IMDC) risk model for RCC in estimating R respectively. Then, we exploited the publicly available datasets of MGS (n=641; NSCLC, RCC and melanoma) to validate TOPOSCORE across different cancer populations. Finally, we developed a friendly user PCR-based score test for SIG bacterial detection allowing the diagnosis of gut dysbiosis within 48 hrs. Table: 259MO

Multivariate analysis for overall survival

Conclusions

TOPOSCORE represents the first easy-to-use and cost-effective tool capable of detecting intestinal dysbiosis associated with longer OS after ICI across cancers on an individual basis. This TOPOSCORE has several implementations, to select donors and recipients of fecal microbial transplantations and follow any microbiota-centered interventions.

Clinical trial identification

NCT04567446.

Legal entity responsible for the study

Gustave Roussy.

Funding

LZ and BB were funded by the RHU Torino Lumière (ANR-16-RHUS-0008). LZ and LD were supported by RHU5 “ANR-21-RHUS-0017” IMMUNOLIFE”. LZ was supported by EU-H2020, Project Number: 825410, Project Acronym: ONCOBIOME, Project title: Gut OncoMicrobiome Signatures (GOMS) associated with cancer incidence, prognosis and prediction of treatment response. LZ also received an ANR grant – French-German Ileobiome - 19-CE15-0029-01. LZ and GK received a donation from Seerave Foundation. LZ, LD were supported by the SIRIC Stratified Oncology Cell DNA Repair and Tumor Immune Elimination (SOCRATE). LD AND LZ was supported by SIGN'IT ARC foundation.

Disclosure

C. Alves Costa Silva: Financial Interests, Personal, Invited Speaker: Bristol Myers Squibb. B. Routy: Financial Interests, Personal and Institutional, Research Grant, outside the submitted work: Davoltera, Kaleido, and Vedanta; Financial Interests, Personal and Institutional, Other: patent for G17004-00006-AD (use of castalagin or analogs thereof for anticancer efficacy and to increase the response to immune-checkpoint inhibitors) pending. B. Escudier: Financial Interests, Personal and Institutional, Advisory Role: AVEO; Bristol Myers Squibb; EUSA Pharma; Ipsen; Novartis; Pfizer; Roche/Genentech; Financial Interests, Institutional, Research Grant: BMS France (Inst); Financial Interests, Personal, Sponsor/Funding, Travel, Accommodations, Expenses: Bristol Myers Squibb; Ipsen; MSD; Pfizer; Roche/Genentech; Financial Interests, Personal, Invited Speaker, Honoraria: Bristol Myers Squibb; EUSA Pharma; Ipsen; Novartis; Oncorena; Pfizer; Roche/Genentech. D. Planchard: Financial Interests, Personal, Invited Speaker, Honoraria: AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, Celgene, Eli Lilly, Merck, Novartis, Pfizer, prIME Oncology, Peer CME, Roche, Janssen, AbbVie; Financial Interests, Personal, Advisory Role, Consulting, advisory role or lectures: AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, Celgene, Daiichi Sankyo, Eli Lilly, Merck, Novartis, Pfizer, prIME Oncology, Peer CME, Roche, Janssen, AbbVie; Financial Interests, Institutional, Principal Investigator, Clinical trials research as principal or co-investigator (Institutional financial interests): AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim, Eli Lilly, Merck, Novartis, Pfizer, Roche, Medimmun, Sanofi-Aventis, Taiho Pharma, Novocure, Daiichi Sankyo, Janssen, AbbVie; Financial Interests, Personal, Sponsor/Funding, Travel, Accommodations, Expenses: AstraZeneca, Roche, Novartis, prIME Oncology, Pfizer. L. Albiges: Financial Interests, Personal and Institutional, Invited Speaker, consulting fees compensated to the institution, outside the submitted work: Pfizer, Novartis, Bristol Myer Squibb, Ipsen, Roche, MSD, AstraZeneca, Merck, Amgen, Astellas, Exelixis, Corvus Pharmaceuticals, Peloton Therapeutics. L. Zitvogel: Financial Interests, Personal, Invited Speaker: Glaxo Smith Kline, Incyte, Lytix, Kaleido, Innovate Pharma, Daiichi Sankyo, Merus, Pilege, Transgene, Tusk, and Roche. All other authors have declared no conflicts of interest.

Background

Lung cancer is one of the top causes of cancer death. Liquid biopsies have emerged as a non-invasive method to identify patients who may respond to therapy, enabling dynamic monitoring of minimal residual disease and drug resistance. While plasma specimens have commonly been used for liquid biopsies, we validated the use of cerebrospinal fluid (CSF) for liquid biopsy next-generation sequencing (NGS) testing.

Methods

31 clinical CSF specimens from patients diagnosed with lung cancer were subjected to NGS testing in a CAP-accredited and CLIA-certified central laboratory using the using the LiquidHALLMARK® assay, a sensitive liquid biopsy NGS test.

Results

Actionable DNA mutations were detected in 77.4% (24/31) of CSF specimens. In 10 cases where CSF and matched plasma were available, the concordance of actionable mutations was 50% (5/10), with actionable mutations found only in CSF in four cases and actionable mutations found only in plasma in one case. Copy number variation (CNV) was detected in 35.5% (11/31) of cases. In five CNV-positive cases where matched plasma was available, CNVs were detected only in CSF and not in matched plasma. This is consistent with the observation by Wang et. al. that more genomic alterations such as CNVs can be detected in CSF than in plasma. Findings from cytology reports using CSF could be obtained in 10 cases (6 positive and 4 negative cases). The concordance between the presence of malignant cells by cytology and actionable variants by NGS was 90% (9/10), with an EGFR exon 19 deletion identified by NGS but no malignant cells detected by cytology in one case. In two further cases, cytology yielded inconclusive results, while NGS reported actionable findings.

Conclusions

These results demonstrate that NGS can sensitively detect the presence of tumor-associated DNA in CSF specimens from lung cancer patients, enabling detection of disease in the central nervous system such as leptomeningeal metastases.

Reference: Wang Y, Jiang F, Xia R, et al. Unique genomic alterations of cerebrospinal fluid cell-free DNA are critical for targeted therapy of non-small cell lung cancer with leptomeningeal metastasis. Front Oncol. Published online October 4, 2021.

Legal entity responsible for the study

Lucence Diagnostics Pte Ltd.

Funding

Lucence Diagnostics Pte Ltd.

Disclosure

M. Tan: Financial Interests, Personal and Institutional, Ownership Interest: Lucence. All other authors have declared no conflicts of interest.

Background

Data on the prevalence of microsatellite instability-high (MSI-H)/deficient mismatch repair (dMMR) in patients with advanced solid tumors from Asia are limited and are needed to aid treatment decisions in the clinical setting. We present real-world data on MSI-H/dMMR across 6 tumor types in Asian patients.

Methods

Patients aged ≥18 y with advanced (stage III/IV) gastrointestinal (GI; biliary tract, gastric, pancreatic) and gynecologic (GYN; cervical, endometrial, ovarian) tumors, regardless of prior therapy, from 29 study centers in Asia-Pacific (Korea, Singapore, Taiwan) and Japan were analyzed. Biomarker testing was performed on archival formalin-fixed paraffin-embedded tissue samples using the Ventana MMR RxDx Panel in patients from Asia-Pacific. Retrospective data on MSI-H PCR status (via MSI-IVD Kit [FALCO]) in patients from Japan were collected. The primary objective was to evaluate the prevalence of MSI-H/dMMR status. Secondary objectives included assessment of clinicopathologic characteristics and treatment patterns.

Results

MSI-H/dMMR status was evaluable in 1970 patients; 980 (49.7%) had GI tumors and 990 (50.3%) had GYN tumors. Prevalence of MSI-H/dMMR was 5.4% (106/1970) in the overall population, 2.8% (27/980) in GI tumors, and 8.0% (79/990) in GYN tumors. Prevalence of MSI-H/dMMR was highest in endometrial cancer in Asia-Pacific (14.6%) and Japan (21.7%) (Table). Most patients in the MSI-H/dMMR population had ECOG PS of 0 or 1 (92.4%) and had undergone surgery (74.5%) and/or received chemotherapy (95.3%) since initial diagnosis. Notably, immune checkpoint inhibitors (ICIs) were administered to a greater proportion of patients with MSI-H/dMMR tumors (34.0%) than to patients with non–MSI-H/proficient MMR tumors (7.2%); most (77.8%) of the MSI-H/dMMR population received ICIs in the second-line or later setting. Table: 261MO

Prevalence of MSI-H/dMMR across tumor types in the overall population and regional cohorts

Conclusions

The prevalence of MSI-H/dMMR across the 6 tumor types in this analysis of Asian patients was consistent with literature reports on non-Asian patients.

Editorial acknowledgement

Medical writing and/or editorial assistance was provided by Obinna T. Ezeokoli, PhD, and Holly C. Cappelli, PhD, CMPP, of ApotheCom (Yardley, PA, USA). This assistance was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.

Legal entity responsible for the study

Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.

Funding

MSD International GmbH (Singapore Branch).

Disclosure

D.S.S.P. Tan: Financial Interests, Personal, Other, Consultancy fees: AstraZeneca, Roche, MSD, Merck Serono, Bayer, Eisai; Financial Interests, Institutional, Research Grant: AstraZeneca, Bayer, Roche, Karyopharm; Financial Interests, Personal, Stocks/Shares: Asian Microbiome Library (AMiLi); Financial Interests, Personal and Institutional, Funding: Singapore Ministry of Health’s National Medical Research Council Clinician Scientist Award and Pangestu Family Foundation Gynaecological Cancer Research Fund. Y.M. Kim: Financial Interests, Institutional, Research Grant: MSD, AstraZeneca, Roche, Regeneron, Clovis Oncology. C. Lu: Financial Interests, Institutional, Principal Investigator: Taichung Veterans General Hospital. M. Kanai: Financial Interests, Personal, Stocks/Shares: Therabiopharma Inc.; Financial Interests, Personal, Advisory Role: Therabiopharma Inc.; Financial Interests, Personal, Speaker’s Bureau: Chugai Pharmaceutical Co., Ltd.; Financial Interests, Institutional, Research Grant: Molecular Health. S.Y. Rha: Other, Personal, Advisory Role: Amgen, Daichii Sankyo, BMS, Indivumed, Merck, BeiGene, Eisai. R. Ramar, M. Wong: Financial Interests, Personal, Full or part-time Employment: MSD International GmbH; Financial Interests, Personal, Stocks/Shares: MSD International GmbH. All other authors have declared no conflicts of interest.

Background

There are nearly half patients with Ras wild-type metastatic colorectal cancer (mCRC) do not response to Anti-EGFR therapy. Identification of patients who are sensitive to anti-EGFR therapy may increase the response rate and reduce the adverse effect. Therefore, there is a pressing need for predicting the efficacy and the clinical benefit in RAS wild type patients. In this study, we aimed to develop and validate a multi-omics deep learning model to predict cetuximab efficacy in RAS wild type mCRC patients.

Methods

In this study, we retrospectively analyzed 213 Ras wild type mCRC patients. Patients in the Arm A (FOLFOX + cetuximab) of CHINESE study (J Clin Oncol 2013, NCT01564810) make up the training set and patients in CHINESE follow-up study (PMID: 30305811) make up the validation set. External multi-omics testing set was derived from an independent cohort and the Arm B (FOLFOX) of the CHINESE study. Based on the deep learning framework Pytorch, we first built the radiomic and genetic signature. Next, we passed the CT images and gene data into the trained ResNet18 and Random Forest and then we sum the output probabilities of two models with a weight of 3:7 to obtain the classification probability of the fusion model.

Results

The signature (area under the ROC curve) successfully predict sensitivity to anti-EGFR therapy (The radiomic signature: 0.63; the genetic signature: 0.72; the fusion signature: 0.81) but failed with chemotherapy (The fusion signature: 0.55). In cetuximab-containing sets, the fusion signature outperformed existing biomarkers for detection of treatment sensitivity and was strongly associated with progression free survival (P<.005).

Conclusions

The multi-omic signature can serve as an intermediate surrogate marker of anti-EGFR treatment sensitivity and survival. The signature outperformed known biomarkers in providing an early prediction of treatment sensitivity and could be used to Ras wild type mCRC treatment decisions.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

EGFR mutant NSCLC comprises the largest molecular subgroup in NSCLC, with remarkable responses to targeted therapy. However a third of the patients do not respond, and among the responders there exists a heterogeneity in response. A better understanding is still needed of the various attributes of the EGFR mutation itself, apart from the widely studied primary resistance mechanisms. Variant Allele frequency (VAF) which is influenced by tumor proportion, and copy number alteration of the gene is an important attribute and a possible predictor of outcome and response.

Methods

126 patients of EGFR mutated NSCLC were subjected to panel based NGS testing. VAF and EGFR copy numbers were recorded and correlated with clinicopathological features and survival outcomes. A VAF of >50% was considered as high VAF, after evaluation by receiver operator curve analysis.

Results

Of 126 EGFR-mutant patients, 67 (53.2%) had exon 19 deletion, while 41 (32.5%) and 18 (14.%) patients had uncommon mutations. EGFR-VAF was significantly higher in patients with exon 19 mutations than in those with exon 21(P<0.003). Exon 19 patients showed a significantly improved PFS (P=0.04) and OS (P=0.07) compared to exon 21 patients. Irrespective of mutation type, a statistically significant association was found between VAF and PFS (P=0.001). High (≥50%) EGFR-VAF was independently associated with a longer PFS [vs. low (<50%) ; median PFS were 12.4 months vs. 9.2 months respectively; P<0.05]. Copy number gain in EGFR more than 5.5 was associated with better PFS of 11.2 months when compared to those with normal copy number. However, copy number gain co-occurring with exon 20 insertion mutation was associated with a shorter PFS (7.2 months vs 12.8 months; p<0.04) when compared to other EGFR subgroups with high copy numbers.

Conclusions

VAF is an important attribute in a panel based NGS report, impacting treatment outcomes and patient survival. This is the first of its kind study from India highlighting the same, and the data on the same from more real world studies are needed for the institution of recommendations for routine reporting of VAF in all such cases.

Legal entity responsible for the study

Ullas Batra.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.