Search

Background Monotherapy immune checkpoint inhibitor (ICI) used in second- or later-line settings has been reported to induce hyperprogression. This study evaluated hyperprogression risk with ICI (atezolizumab) in the first-, second-, or later-line treatment of advanced non–small cell lung cancer (NSCLC), and provides insights into hyperprogression risk with contemporary first-line ICI treatment. Methods Hyperprogression was identified using Response Evaluation Criteria in Solid Tumours (RECIST)-based criteria in a dataset of pooled individual-participant level data from BIRCH, FIR, IMpower130, IMpower131, IMpower150, OAK, and POPLAR trials. Odds ratios were computed to compare hyperprogression risks between groups. Landmark Cox proportional-hazard regression was used to evaluate the association between hyperprogression and progression-free survival/overall survival. Secondarily, putative risk factors for hyperprogression among second- or later-line atezolizumab-treated patients were evaluated using univariate logistic regression models. Results Of the included 4644 patients, 119 of the atezolizumab-treated patients (n = 3129) experienced hyperprogression. Hyperprogression risk was markedly lower with first-line atezolizumab—either chemoimmunotherapy or monotherapy—compared to second/later-line atezolizumab monotherapy (0.7% vs. 8.8%, OR = 0.07, 95% CI, 0.04-0.13). Further, there was no statistically significant difference in hyperprogression risk with first-line atezolizumab-chemoimmunotherapy versus chemotherapy alone (0.6% vs. 1.0%, OR = 0.55, 95% CI, 0.22-1.36). Sensitivity analyses using an extended RECIST-based criteria including early death supported these findings. Hyperprogression was associated with worsened overall survival (HR = 3.4, 95% CI, 2.7-4.2, P < .001); elevated neutrophil-to-lymphocyte ratio was the strongest risk factor for hyperprogression (C-statistic = 0.62, P < .001). Conclusions This study presents first evidence for a markedly lower hyperprogression risk in advanced NSCLC patients treated with first-line ICI, particularly with chemoimmunotherapy, as compared to second- or later-line ICI treatment.

Immunogenicity; Proteogenomics Inmunogenicidad; Proteogenómica Immunogenicitat; Proteogenòmica Purpose: Tumor antigens are central to antitumor immunity. Recent evidence suggests that peptides from noncanonical (nonC) aberrantly translated proteins can be presented on HLA-I by tumor cells. Here, we investigated the immunogenicity of nonC tumor HLA-I ligands (nonC-TL) to better understand their contribution to cancer immunosurveillance and their therapeutic applicability. Experimental Design: Peptides presented on HLA-I were identified in 9 patient-derived tumor cell lines from melanoma, gynecologic, and head and neck cancer through proteogenomics. A total of 507 candidate tumor antigens, including nonC-TL, neoantigens, cancer-germline, or melanocyte differentiation antigens, were tested for T-cell recognition of preexisting responses in patients with cancer. Donor peripheral blood lymphocytes (PBL) were in vitro sensitized against 170 selected nonC-TL to isolate antigen-specific T-cell receptors (TCR) and evaluate their therapeutic potential. Results: We found no recognition of the 507 nonC-TL tested by autologous ex vivo expanded tumor-reactive T-cell cultures while the same cultures demonstrated reactivity to mutated, cancer-germline, or melanocyte differentiation antigens. However, in vitro sensitization of donor PBL against 170 selected nonC-TL, led to the identification of TCRs specific to three nonC-TL, two of which mapped to the 5′ UTR regions of HOXC13 and ZKSCAN1, and one mapping to a noncoding spliced variant of C5orf22C. T cells targeting these nonC-TL recognized cancer cell lines naturally presenting their corresponding antigens. Expression of the three immunogenic nonC-TL was shared across tumor types and barely or not detected in normal cells. Conclusions: Our findings predict a limited contribution of nonC-TL to cancer immunosurveillance but demonstrate they may be attractive novel targets for widely applicable immunotherapies. We thank the patients for their participation in this study, Steven A. Rosenberg for providing valuable reagents and support for NGS studies, R. Pujol for helpful scientific discussion, J. Gonzalez for bioinformatics support, CRG/UPF Flow Cytometry Unit for assistance with cell sorting, and CRG/UPF and IRB Proteomics Units for technical support. A. Gros and this work were funded by the Comprehensive Program of Cancer Immunotherapy & Immunology II (CAIMI-II) supported by the BBVA Foundation (53/2021), Institute Carlos III (MS15/00058 and PI17/01085), AECC (IDEAS197PORT), and La Fundació La Marató de TV3 (201919–30). We thank CERCA Programme / Generalitat de Catalunya for institutional support. M. Lozano-Rabella was supported by the Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR) (2018FI_B 00946). A. Garcia-Garijo was supported by Generalitat PERIS award (SLT017/20/000131). A. Yuste-Estevanez was supported by the Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR) (2021 FI_B 00365). J. Palomero was supported by the Beatriu de Pinós programme (BP 2018), cofounded by the Agency for Management of University and Research Grants (AGAUR) and European Union's Horizon 2020.

Table 1: samples included in the RNAseq analysis Figure 1: consort flow diagram Table 2: Comparison of patients' characteristics of the subset of patients included in the RNA-seq analysis and patients recruited in the DECISION study Figure 2: Progression-free survival of the subgroup of patients included in the RNA-seq analysis. Table 3: Pathways_GO_analysis Figure 3: Kaplan-Meyer curves of the three RNA expression groups regarding treatment group in DECISION trial. Figure 4: A,B,C: PFS in papillary, follicular and poorly differentiated histologies, respectively; D,E,F: OS in papillary, follicular and poorly differentiated histologies, respectively Figure 5: Overall survival (OS) analysis based on RNA expression profile from TCGA data of papillary thyroid cancer (Cell, 2014) Figure 6: Analyses based on 121 angiogenesisrelated genes

Several biomarkers have been suggested to have prognostic value in differentiated thyroid carcinomas (DTC) with no validation in the refractory setting, including all tumor subtypes. We aim to correlate RNA expression profiles with survival based on patients included in the DECISION trial. We obtained 247 samples from the 417 patients included in the DECISION study and performed RNAseq analysis (77 million paired-end reads for each sample on HiSeq2000). After quality control, 125 samples were included in the secondary analysis and mapped against the human reference genome (GRCh38) with STAR (v2.5.1b) using ENCODE parameter. Survival analysis was calculated using the Kaplan–Meier method and log-rank test was used for statistical comparison. In this post hoc analysis, we identified three groups of tumors based on their gene expression profile: BRAF-like, RAS-like, and non-BRAF-non-RAS-like (NoBRaL). No significant correlation with sorafenib responders was observed. However, we identified a statistically significant correlation between the RNA-expression profiles and progression-free survival. The BRAF-like profile had a significantly better outcome compared with RAS-like and NoBRaL (11.8, 6.2, and 5.5 months, respectively) [HR: 0.31, 95% confidence interval (CI), 0.17–0.60; P < 0.001 and HR: 0.36 (95% CI, 0.21–0.63); P < 0.001] and HR: 0.36 (95% CI, 0.21–0.63; P < 0.001) and maintained significance as an independent prognostic factor for overall survival in the multivariate analysis for papillary thyroid cancers. To our knowledge, this is the first comprehensive RNA-seq analysis of all histologic subtypes of DTC. The RNA expression profiles identified may suggest a new prognostic parameter to be considered before recommendation of systemic therapies or the design of stratification factors for future clinical trials.

Includes a large list of all the peptides and details of the HLA-I ligands identified at =30 ALC score

Supplementary Figure 1. Validation of HLA-I ligand tumor antigen candidates with synthetic peptides Supplementary Figure 2. Validation of selected HLA-I ligand tumor antigen candidates identified in Mel-1 and Mel-3 with isotope-labeled peptides Supplementary Figure 3. Frequency of nonC peptides detected in HLA-I. Supplementary Fig. 4. HLA-I allele binding motif of all patients included in the study. Supplementary Figure 5. A fraction of nonC-TL can be detected in publicly available immunopeptidomics datasets from human tumor biopsies Supplementary Figure 6. Evaluation of pre-existing T-cell responses to candidate tumor antigens in cancer patients Supplementary Figure 7. Enrichment of antigen-specific T cells from cancer patients by flow cytometry-based sorting. Supplementary Figure 8. Deconvolution of peptide pools containing recognized CGA or melanoma-associated antigens. Supplementary Figure 9. Functional avidity and specificity of neoantigen-specific T cells isolated from cancer patients. Supplementary Figure 10. HLA restriction element and recognition of autologous tumor cell lines by antigen-specific T cells isolated from cancer patients. SupplementaryFigure 11. Genomic location of the three immunogenic nonC-TL identified through IVS. Supplementary Figure 12. Recognition of the predicted ORFs encoding the immunogenic nonC-TL by TCR transduced cells. Supplementary Figure 13. Loss of recognition of TCL transfected with Cas9-sgRNA targeting the predicted 5’U-HOXC13 ORF. Supplementary Figure 14. Assessment of expression and translation of the immunogenic nonC-TL in human healthy cells. Supplementary Figure 15. NonC-TL are the main source of candidate tumor antigens identified by proteogenomics but tumor-reactive T cells preferentially recognize neoantigens derived from NSM.

Show patient characteristics and relevant experimental information

Supplementary Table 2

Supplementary Table 1

Supplementary Figure 2

Supplementary Table 4

Figure S1. Flowchart of study selection process with ICIs treatment. Figure S2. Flowchart of study selection process with TAs treatment. Figure S3. Overall Survival comparing HPD with non-HPD in patients with progression disease as best response in the ICIs and TAs cohorts ( RECIST 1.1 progression disease Landmark Analysis). Figure S4. Overall Survival in second line (a) and third (b) setting HPD vs non-HPD progressors by RECIST criteria in ICIs cohort. Figure S5. Concordance between RECIST and TGR criteria in ICIs cohort. Table S1. Computed tomography (CT) acquisition protocol parameters. Table S2. Clinical variables in patients treated with TAs. Table S3. Multivariate model in patients with progression disease as best response in ICIs group (using TGR).

Supplementary Table 3

Supplementary Figure 1

Supplementary Figure Legends

Purpose:Most hyperprogression disease (HPD) definitions are based on tumor growth rate (TGR). However, there is still no consensus on how to evaluate this phenomenon.Patients and Methods:We investigated two independent cohorts of patients with advanced solid tumors treated in phase I trials with (i) programmed cell death 1 (PD-1)/PD-L1 antibodies in monotherapy or combination and (ii) targeted agents (TA) in unapproved indications. A Response Evaluation Criteria in Solid Tumors (RECIST) 1.1–based definition of hyperprogression was developed. The primary endpoint was the assessment of the rate of HPD in patients treated with ICIs or TAs using both criteria (RECIST and TGR) and the impact on overall survival (OS) in patients who achieved PD as best response.Results:Among 270 evaluable patients treated with PD-1/PD-L1 inhibitors, 29 PD-1/PD-L1–treated patients (10.7%) had HPD by RECIST definition. This group had a significantly lower OS (median of 5.23 months; 95% CI, 3.97–6.45) when compared with the non-HPD progressor group (median, 7.33 months; 95% CI, 4.53–10.12; HR = 1.73, 95% CI, 1.05–2.85; P = 0.04). In a subset of 221 evaluable patients, 14 (6.3%) were categorized as HPD using TGR criteria, differences in median OS (mOS) between this group (mOS 4.2 months; 95% IC, 2.07–6.33) and non-HPD progressors (n = 44) by TGR criteria (mOS 6.27 months; 95% CI, 3.88–8.67) were not statistically significant (HR 1.4, 95% IC, 0.70–2.77; P = 0.346). Among 239 evaluable patients treated with TAs, 26 (10.9%) were classified as having HPD by RECIST and 14 using TGR criteria in a subset of patients. No differences in OS were observed between HPD and non-HPD progressors treated with TAs.Conclusions:HPD measured by TGR or by RECIST was observed in both cohorts of patients; however, in our series, there was an impact on survival only in the immune-checkpoint inhibitor cohort when evaluated by RECIST. We propose a new way to capture HPD using RECIST criteria that is intuitive and easy to use in daily clinical practice.

Supplementary Figure 3

Purpose:The limited knowledge of the molecular alterations that characterize poorly differentiated neuroendocrine carcinomas has limited the clinical development of targeted agents directed to driver mutations. Here we aim to identify new molecular targets in colon neuroendocrine carcinomas (co-NEC) and proof the efficacy of matching drugs.Experimental Design:We performed a multi-omic analysis of co-NEC to identify genetic or epigenetic alterations that could be exploited as effective drug targets. We compared co-NEC samples with colorectal carcinomas (CRC) to identify neuroendocrine-specific traits. Patients with co-NEC and patient-derived xenografts were treated with a BRAFV600E-blocking drug to demonstrate sensitivity.Results:co-NEC and CRC are similar in their mutational repertoire, although co-NECs are particularly enriched in BRAFV600E mutations. We report for the first time that V600EBRAF-mutant co-NECs may benefit from BRAF inhibition in monotherapy and how EGFR status is essential to predict innate sensitivity and acquired resistance by a differential methylation of its gene regulatory regions.Conclusions:The identification of V600E BRAF mutations in high-grade co-NECs has allowed the description of radiological responses to combination therapy of BRAF and MEK inhibitors in basket clinical trials. However, the molecular rationale for this treatment combination was based on the presence of the BRAF mutation and the efficacy observed in other cancer types such as melanoma. Future drug development in this setting should test BRAF inhibitors upfront and the addition of anti-EGFR antibodies instead of MEK inhibitors for an efficient blockade of acquired resistance.

Tumor antigens are central to antitumor immunity. Recent evidence suggests that peptides from non-canonical (nonC) aberrantly translated proteins can be presented on HLA-I by tumor cells. Here, we investigated the immunogenicity of nonC tumor HLA-I ligands (nonC-TL) to better understand their contribution to cancer immunosurveillance and their therapeutic applicability. Using proteogenomics, we identified 517 nonC-TL from 9 patients with melanoma, gynecological, and head and neck cancer. We found no recognition of the 507 nonC-TL tested by autologousex vivoexpanded tumor reactive T-cell cultures while the same cultures demonstrated reactivity to mutated, cancer-germline, or melanocyte differentiation antigens. However,in vitrosensitization of donor peripheral blood lymphocytes against 170 selected nonC-TL, led to the identification of T-cell receptors (TCRs) specific to three nonC-TL, two of which mapped to the 5’ UTR regions of HOXC13 and ZKSCAN1, and one mapping to a non-coding spliced variant of C5orf22C. T cells targeting these nonC-TL recognized cancer cell lines naturally presenting their corresponding antigens. Expression of the three immunogenic nonC-TL was shared across tumor types and barely or not detected in normal cells. Our findings predict a limited contribution of nonC-TL to cancer immunosurveillance but demonstrate they may be attractive novel targets for widely applicable immunotherapies.

Background Reliable predictive imaging markers of response to immune checkpoint inhibitors are needed. Purpose To develop and validate a pretreatment CT-based radiomics signature to predict response to immune checkpoint inhibitors in advanced solid tumors. Materials and Methods In this retrospective study, a radiomics signature was developed in patients with advanced solid tumors (including breast, cervix, gastrointestinal) treated with anti-programmed cell death-1 or programmed cell death ligand-1 monotherapy from August 2012 to May 2018 (cohort 1). This was tested in patients with bladder and lung cancer (cohorts 2 and 3). Radiomics variables were extracted from all metastases delineated at pretreatment CT and selected by using an elastic-net model. A regression model combined radiomics and clinical variables with response as the end point. Biologic validation of the radiomics score with RNA profiling of cytotoxic cells (cohort 4) was assessed with Mann-Whitney analysis. Results The radiomics signature was developed in 85 patients (cohort 1: mean age, 58 years ± 13 [standard deviation]; 43 men) and tested on 46 patients (cohort 2: mean age, 70 years ± 12; 37 men) and 47 patients (cohort 3: mean age, 64 years ± 11; 40 men). Biologic validation was performed in a further cohort of 20 patients (cohort 4: mean age, 60 years ± 13; 14 men). The radiomics signature was associated with clinical response to immune checkpoint inhibitors (area under the curve [AUC], 0.70; 95% CI: 0.64, 0.77

Purpose: The limited knowledge of the molecular alterations that characterize poorly differentiated neuroendocrine carcinomas has limited the clinical development of targeted agents directed to driver mutations. Here we aim to identify new molecular targets in colon neuroendocrine carcinomas (co-NEC) and proof the efficacy of matching drugs. Experimental Design: We performed a multi-omic analysis of co-NEC to identify genetic or epigenetic alterations that could be exploited as effective drug targets. We compared co-NEC samples with colorectal carcinomas (CRC) to identify neuroendocrine-specific traits. Patients with co-NEC and patient-derived xenografts were treated with a BRAFV600E-blocking drug to demonstrate sensitivity. Results: co-NEC and CRC are similar in their mutational repertoire, although co-NECs are particularly enriched in BRAFV600E mutations. We report for the first time that V600EBRAF-mutant co-NECs may benefit from BRAF inhibition in monotherapy and how EGFR status is essential to predict innate sensitivity and acquired resistance by a differential methylation of its gene regulatory regions. Conclusions: The identification of V600E BRAF mutations in high-grade co-NECs has allowed the description of radiological responses to combination therapy of BRAF and MEK inhibitors in basket clinical trials. However, the molecular rationale for this treatment combination was based on the presence of the BRAF mutation and the efficacy observed in other cancer types such as melanoma. Future drug development in this setting should test BRAF inhibitors upfront and the addition of anti-EGFR antibodies instead of MEK inhibitors for an efficient blockade of acquired resistance.

Several biomarkers have been suggested to have prognostic value in differentiated thyroid carcinomas (DTC) with no validation in the refractory setting, including all tumor subtypes. We aim to correlate RNA expression profiles with survival based on patients included in the DECISION trial. We obtained 247 samples from the 417 patients included in the DECISION study and performed RNAseq analysis (77 million paired-end reads for each sample on HiSeq2000). After quality control, 125 samples were included in the secondary analysis and mapped against the human reference genome (GRCh38) with STAR (v2.5.1b) using ENCODE parameter. Survival analysis was calculated using the Kaplan–Meier method and log-rank test was used for statistical comparison. In this post hoc analysis, we identified three groups of tumors based on their gene expression profile: BRAF-like, RAS-like, and non-BRAF-non-RAS-like (NoBRaL). No significant correlation with sorafenib responders was observed. However, we identified a statistically significant correlation between the RNA-expression profiles and progression-free survival. The BRAF-like profile had a significantly better outcome compared with RAS-like and NoBRaL (11.8, 6.2, and 5.5 months, respectively) [HR: 0.31, 95% confidence interval (CI), 0.17–0.60; P < 0.001 and HR: 0.36 (95% CI, 0.21–0.63); P < 0.001] and HR: 0.36 (95% CI, 0.21–0.63; P < 0.001) and maintained significance as an independent prognostic factor for overall survival in the multivariate analysis for papillary thyroid cancers. To our knowledge, this is the first comprehensive RNA-seq analysis of all histologic subtypes of DTC. The RNA expression profiles identified may suggest a new prognostic parameter to be considered before recommendation of systemic therapies or the design of stratification factors for future clinical trials.

BackgroundDespite recent advances in exome and RNA sequencing to identify tumor-rejection antigens including neoantigens, the existing techniques fail to identify the vast majority of antigens targeted by tumor-reactive cells. A growing number of studies suggest that HLA-I peptides derived from non-canonical (nonC) open reading frames or derived from allegedly non-coding regions can contribute to tumor immunogenicity. Here we use proteogenomics to identify personalized candidate canonical and non-canonical tumor-rejection antigens and to evaluate their contribution to cancer immune surveillance in patients.MethodsWhole exome sequencing was performed to identify the non-synonymous somatic mutations (NSM) and immunopeptidomics to identify the HLA-I presented peptides (pHLA) in 9 patient-derived tumor cell lines (TCL). Peptid-PRISM proteogenomics pipeline was used to identify both canonical and non-canonical pHLA, including those derived from NSM in coding regions. All peptides containing mutations and derived from either cancer-testis (CTA) or tumor-associated antigens (TAA) were selected as candidate tumor antigens. For nonC peptides, an immunopeptidomics healthy dataset containing several tissues and HLA-allotypes was used to eliminate those derived from normal ORFs and select nonC peptides preferentially expressed in tumor cells (nonC-TE). The selected candidate peptides were synthesized, pulsed onto autologous APCs and co-cultured with tumor-reactive ex vivo expanded lymphocytes to assess immune recognition (figure 1).ResultsNonC-TE peptides were identified in all TCL studied, ranging from 0.5% to 5.4% of the total HLA-I presented peptides (n= 506). As described previoulsy, 5’UTR were the main source. Of note, the tumor type did not have an impact on the frequency of presented nonC peptides, but rather the presence of HLA-A*11:01 and HLA-A*03:01 was a major determinant. T cell responses were detected against at least 13/33 putative neoantigens, 2/24 CTA and 2/61 TAA. On the contrary, none of the 471 nonC-TE candidate peptides tested thus far, including one containing a NSM were able to elicit a recall immune response. Nevertheless, T cells recognizing at least 3 of them were detected through in vitro sensitization of non-autologous PBMCs.Abstract 912 Figure 1Workflow diagramTumor biopsies and blood samples are obtained from cancer patients (left panel). Patient-derived tumor cell lines are generated in vitro, the peptides presented on HLA molecules are further isolated and analyzed in a mass-spectrometer (top panel). Whole exome sequencing (WES) from matched tumor and healthy tissue is performed to identify the non-synonymous somatic mutations (NSM) (middle panel). Peptide-PRISM proteogenomics pipeline combines the information from the immunopeptidomics data and WES to identify pHLA sequences from both canonical and non-canonical candidate tumor antigens (top right panel). Lymphocyte populations either TILs or sorted PBMCs are expanded and further screened for pre-existing T cell responses (bottom panel) against the candidate epitopes by co-culturing the T cells with peptide-pulsed autologous APC. The recognition is assessed by measuring IFNg release by elispot and the upregulation of activation surface markers by FACS (bottom right panel).ConclusionsOur results show that although HLA-I nonC peptides were frequently presented in all TCLs studied and they can be immunogenic, neoantigens derived from mutations in canonical coding regions were preferentially recognized by tumor-reactive lymphocytes, suggesting T cells targeting the latter are primed more efficiently. The identification of mutated nonC antigens using whole genome sequencing to identify mutations in non-coding regions warrants further examination. Still, the specificity of many tumor-reactive TILs remains unknown.Ethics Approval”This study was approved by the ”Comité de Ética de Investigación con Medicamentos del Hospital Universitario Vall d’Hebron” institution’s Ethics Board; approval number PR(AG)537/2019.”

Despite the clinical success of checkpoint inhibitors, a substantial gap still exists in our understanding of their mechanism of action. While antibodies to cytotoxic T lymphocyte-associated protein-4 (CTLA-4) were developed to block inhibitory signals in T cells, several recent studies have demonstrated that Fcγ receptor (FcγR)-dependent depletion of regulatory T cells (T

Immunotherapy; Phase 1 trials; Prognostic model Inmunoterapia; Ensayos de fase 1; Modelo pronóstico Immunoteràpia; Assajos de fase 1; Model pronòstic Purpose Patient selection in phase 1 clinical trials (Ph1t) continues to be a challenge. The aim of this study was to develop a user-friendly prognostic calculator for predicting overall survival (OS) outcomes in patients to be included in Ph1t with immune checkpoint inhibitors (ICIs) or targeted agents (TAs) based on clinical parameters assessed at baseline. Methods Using a training cohort with consecutive patients from the VHIO phase 1 unit, we constructed a prognostic model to predict median OS (mOS) as a primary endpoint and 3-month (3m) OS rate as a secondary endpoint. The model was validated in an internal cohort after temporal data splitting and represented as a web application. Results We recruited 799 patients (training and validation sets, 558 and 241, respectively). Median follow-up was 21.2 months (m), mOS was 10.2 m (95% CI, 9.3–12.7) for ICIs cohort and 7.7 m (95% CI, 6.6–8.6) for TAs cohort. In the multivariable analysis, six prognostic variables were independently associated with OS – ECOG, number of metastatic sites, presence of liver metastases, derived neutrophils/(leukocytes minus neutrophils) ratio [dNLR], albumin and lactate dehydrogenase (LDH) levels. The phase 1 prognostic online (PIPO) calculator showed adequate discrimination and calibration performance for OS, with C-statistics of 0.71 (95% CI 0.64–0.78) in the validation set. The overall accuracy of the model for 3m OS prediction was 87.2% (95% CI 85%–90%). Conclusions PIPO is a user-friendly objective and interactive tool to calculate specific survival probabilities for each patient before enrolment in a Ph1t. The tool is available at https://pipo.vhio.net/. The research leading to these results has received funding from “la Caixa” Foundation (LCF/PR/CE07/50610001). Cellex Foundation for providing research facilities and equipment. This work was supported by the Accelerator Award (UpSMART) from Fundacion Científica – Asociacion Espanola Contra el Cancer (FC -AECC)/ Associazione Italiana per la Ricerca sul Cancro (AIRC) /Cancer Research United Kingdom (CRUK).

PURPOSEApproved systemic therapies for advanced gastroenteropancreatic neuroendocrine tumors (GEP-NETs) have shown limited capacity to reduce tumor burden and no antitumor activity after progression to targeted agents (TAs). We investigated the efficacy and safety of lenvatinib in patients with previously treated advanced GEP-NETs.PATIENTS AND METHODSThis was a multicenter, single-arm, open-label, phase II trial with two parallel cohorts (ClinicalTrials.gov identifier: NCT02678780 ) involving 21 institutions in 4 European countries. Eligible patients had histologically confirmed advanced grade 1-2 pancreatic (panNET) or GI (GI-NET) NETs with documented tumor progression after treatment with a TA (panNET) or somatostatin analogs (GI-NET). Patients were treated with lenvatinib 24 mg once daily until disease progression or treatment intolerance. The primary end point was overall response rate by central radiology review. Secondary end points included progression-free survival, overall survival, duration of response, and safety.RESULTSBetween September 2015 and March 2017, a total of 111 patients were enrolled, with 55 (panNET) and 56 (GI-NET) patients in each cohort. The median follow-up was 23 months. The overall response rate was 29.9% (95% CI, 21.6 to 39.6): 44.2% (panNET) and 16.4% (GI-NET). The median (range) duration of response was 19.9 (8.4-30.8) and 33.9 (10.6-38.3) months in the panNET and GI-NET groups, respectively. The median progression-free survival was 15.7 months (95% CI, 14.1 to 19.5). The most common adverse events were fatigue, hypertension, and diarrhea; 93.7% of patients required dose reductions or interruptions.CONCLUSIONWe report the highest centrally confirmed response reported to date with a multikinase inhibitor in advanced GEP-NETs, with a particularly strong response in the panNET cohort. This study provides novel evidence for the efficacy of lenvatinib in patients with disease progression following treatment with other TAs, suggesting the potential value of lenvatinib in the treatment of advanced GEP-NETs.

Over the last several years, several molecular aberrations have been unevenly described across cancers, although the distinct functional relevance in each biological context is not yet fully understood. Novel discoveries have led to the development of drugs tailored to the molecular profile of patients, thus increasing the likelihood of response among biomarker-selected patients. In this context, there has been a progressive redefinition of a precision medicine framework where evidence-based development and earlier approvals might now be driven by this molecular information. Innovative trial designs have greatly facilitated the evaluation and approval of new drugs in small cohorts of orphan cancers in which histology-dependent molecularly defined trials might be logistically difficult. However, accelerated approvals based on this agnostic-histology development model have brought new clinical, regulatory, and reimbursement challenges. In this article, we will highlight many of the biologic issues and clinical trial design challenges characterizing the development of tissue-agnostic compounds. Also, we will review some of the key factors involved in the development of pembrolizumab and larotrectinib, the first two drugs that have been approved by the U.S. Food and Drug Administration in an histology-agnostic manner. Because we anticipate that agnostic-histology approvals will continue to grow, we aim to provide insight into the current panorama of targeted drugs that are following this strategy and some premises to take into consideration. Clinicians and regulators should be prepared to overcome the associated potential hurdles, ensuring that uncertainties are dealt with properly and allowing new, promising agents to arrive faster to the market.

INTRODUCTION The addition of monoclonal antibody (mAb) epidermal growth factor receptor (EGFR) inhibitors to classic chemotherapy doublet backbones has improved survival of metastatic colorectal cancer (mCRC). However, the role of triple-drug chemotherapy regimens in combination with an anti-EGFR mAb inhibitor is not yet clear. AREAS COVERED The activity of triple-drug chemotherapy regimens when combined with an anti-EGFR mAb in mCRC patients is examined. We describe the overall safety and tolerability profiles based on a literature review of all published phase I and II clinical trials in this setting. Drug exposure, tumor mutational status, and metastases resectability are discussed. A review of PubMed and abstracts of major oncology congresses from 2009 to 2018, with MeSH and full-text search terms for clinical trials of anti-EGFR for 'metastatic' or 'advanced' 'colorectal cancer/adenocarcinoma' was implemented. Only English language publications were included. EXPERT OPINION Efficacy data from phase II trials are promising, but the safety profiles are not as encouraging; the development of severe diarrhea and acneiform rash limit the drug exposure that is critical for improved outcomes. Phase II studies of these triplet chemotherapy/anti-EGFR mAb combinations have focused on conversion therapy in liver-limited disease or in the first-line setting in advanced disease. The identification of biomarkers of response and toxicity may support the use of personalized medicine and more precise design of phase III trials.

Summary Background Oncogenic alterations in RET represent important therapeutic targets in thyroid cancer. We aimed to assess the safety and antitumour activity of pralsetinib, a highly potent, selective RET inhibitor, in patients with RET-altered thyroid cancers. Methods ARROW, a phase 1/2, open-label study done in 13 countries across 71 sites in community and hospital settings, enrolled patients 18 years or older with RET-altered locally advanced or metastatic solid tumours, including RET-mutant medullary thyroid and RET fusion-positive thyroid cancers, and an Eastern Co-operative Oncology Group performance status of 0–2 (later limited to 0–1 in a protocol amendment). Phase 2 primary endpoints assessed for patients who received 400 mg once-daily oral pralsetinib until disease progression, intolerance, withdrawal of consent, or investigator decision, were overall response rate (Response Evaluation Criteria in Solid Tumours version 1.1; masked independent central review) and safety. Tumour response was assessed for patients with RET-mutant medullary thyroid cancer who had received previous cabozantinib or vandetanib, or both, or were ineligible for standard therapy and patients with previously treated RET fusion-positive thyroid cancer; safety was assessed for all patients with RET-altered thyroid cancer. This ongoing study is registered with clinicaltrials.gov , NCT03037385 , and enrolment of patients with RET fusion-positive thyroid cancer was ongoing at the time of this interim analysis. Findings Between Mar 17, 2017, and May 22, 2020, 122 patients with RET-mutant medullary and 20 with RET fusion–positive thyroid cancers were enrolled. Among patients with baseline measurable disease who received pralsetinib by July 11, 2019 (enrolment cutoff for efficacy analysis), overall response rates were 15 (71%) of 21 (95% CI 48–89) in patients with treatment-naive RET-mutant medullary thyroid cancer and 33 (60%) of 55 (95% CI 46–73) in patients who had previously received cabozantinib or vandetanib, or both, and eight (89%) of nine (95% CI 52–100) in patients with RET fusion-positive thyroid cancer (all responses confirmed for each group). Common (≥10%) grade 3 and above treatment-related adverse events among patients with RET-altered thyroid cancer enrolled by May 22, 2020, were hypertension (24 patients [17%] of 142), neutropenia (19 [13%]), lymphopenia (17 [12%]), and anaemia (14 [10%]). Serious treatment-related adverse events were reported in 21 patients (15%), the most frequent (≥2%) of which was pneumonitis (five patients [4%]). Five patients [4%] discontinued owing to treatment-related events. One (1%) patient died owing to a treatment-related adverse event. Interpretation Pralsetinib is a new, well-tolerated, potent once-daily oral treatment option for patients with RET-altered thyroid cancer. Funding Blueprint Medicines.

Summary Background Malignant pleural mesothelioma (MPM) has few treatment options. Pembrolizumab showed preliminary clinical benefit in programmed death ligand 1 (PD-L1)-positive MPM. We evaluated the efficacy and safety of pembrolizumab monotherapy in patients with previously treated MPM irrespective of PD-L1 status in the KEYNOTE-158 study. Methods The ongoing open-label, multicohort, single-arm, phase 2 KEYNOTE-158 study enrolled eligible adults (≥18 years) with MPM who had progression on or intolerance to standard therapy, Eastern Cooperative Oncology Group performance status 0–1, and biomarker-evaluable tumour samples. Individuals were enrolled from 35 academic facilities and community-based institutions across 14 countries in Australia, North America, Europe, and Asia. Participants received pembrolizumab 200 mg intravenously every 3 weeks for up to 35 cycles. The primary efficacy endpoint was objective response per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, based on radiological imaging every 9 weeks for the first year of the study and every 12 weeks thereafter and assessed by independent central review. Efficacy and safety were analysed in all patients who received at least one dose of pembrolizumab. This trial is registered with ClinicalTrials.gov , NCT02628067 . Findings Patients were enrolled in the MPM cohort between Feb 9, 2016, and Aug 16, 2016. As of June 27, 2019, 118 patients had been enrolled and received at least one dose of pembrolizumab. Ten (8% [95% CI 4–15]) patients had an objective response. Median duration of objective response was 14·3 months (range 4·0 to 33·9+), and 60% of objective responses were ongoing at 12 months. Objective responses were observed in six (8%) of 77 patients with PD-L1-positive MPM (median response duration 17·7 months [range 5·8 to 33·9+]) and four (13%) of 31 patients with PD-L1-negative MPM (10·2 months [4·0–16·6]). Median overall survival was 10·0 months (95% CI 7·6–13·4) and median progression-free survival was 2·1 months (2·1–3·9). Treatment-related adverse events occurred in 82 (69%) of 118 patients and serious adverse events that were considered to be treatment-related occurred in 14 (12%) of 118 patients. 19 (16%) patients had grade 3–4 treatment-related events, and most common of these were colitis (three patients), hyponatraemia (three), and pneumonitis (two). One patient died from treatment-related apnoea. By the end of the trial, 113 (96%) patients had discontinued pembrolizumab and progressive disease was the most common reason for discontinuation. Interpretation Pembrolizumab showed durable antitumour activity and manageable toxicity in patients with advanced MPM, regardless of PD-L1 status. Our data support the programmed death 1 (PD-1) and PD-L1 pathway as a potential therapeutic target in some patients with previously treated mesothelioma but biomarkers that can effectively identify such patients are yet to be elucidated. Funding Merck Sharp & Dohme Corp, a subsidiary of Merck & Co, Inc, Kenilworth, NJ, USA.

In the immunoncology era, an acceleration of tumor growth upon immune checkpoint inhibitors (ICI), defined as hyperprogressive disease (HPD) has been observed across different cancers. Although in non-small-cell lung cancer, most of the available evidence regarding HPD has been reported for patients treated with single agent PD-1 and PD-L1 inhibitors, in retrospective series a variable proportion of patients receiving ICI combinations also experienced HPD. Similarly, the shape of survival curves and the progression rates in clinical trials testing combinations of PD-1/PD-L1 inhibitors and anti-CTLA-4 agents suggest the occurrence of HPD. Few data are available regarding pseudoprogression upon ICI combinations. However, considering that pseudoprogression has been reported for anti-PD-1/PD-L1 agents and for CTLA-4 inhibitors separately, it is likely that it may occur also upon combinations of these two classes of drugs.

Background Futibatinib is an oral, irreversible, highly selective fibroblast growth factor receptor (FGFR)1–4 inhibitor with potent preclinical activity against tumors harboring FGFR aberrations. This first-in-human, phase I dose-escalation trial (NCT02052778) evaluates the safety and pharmacokinetics/pharmacodynamics of futibatinib in advanced solid tumors. Patients and methods Following a standard 3+3 dose-escalation design, eligible patients with advanced solid tumors refractory to standard therapies received 8–200 mg futibatinib three times a week (t.i.w.) or 4–24 mg once daily (q.d.). Results A total of 86 patients were enrolled in the nine t.i.w. (n = 42) and five q.d. cohorts (n = 44); 71 patients (83%) had tumors harboring FGF/FGFR aberrations. Three of nine patients in the 24-mg q.d. cohort experienced dose-limiting toxicities, including grade 3 increases in alanine transaminase, aspartate transaminase, and blood bilirubin (n = 1 each). The maximum tolerated dose (MTD) was determined to be 20 mg q.d.; no MTD was defined for the t.i.w. schedule. Across cohorts (n = 86), the most common treatment-emergent adverse events (TEAEs) were hyperphosphatemia (59%), diarrhea (37%), and constipation (34%); 48% experienced grade 3 TEAEs. TEAEs led to dose interruptions, dose reductions, and treatment discontinuations in 55%, 14%, and 3% of patients, respectively. Pharmacokinetics were dose proportional across all q.d. doses but not all t.i.w. doses evaluated, with saturation observed between 80 and 200 mg t.i.w. Serum phosphorus increased dose dependently with futibatinib on both schedules, but a stronger exposure–response relationship was observed with q.d. dosing, supporting 20 mg q.d. as the recommended phase II dose (RP2D). Overall, partial responses were observed in five patients [FGFR2 fusion-positive intrahepatic cholangiocarcinoma (n = 3) and FGFR1-mutant primary brain tumor (n = 2)], and stable disease in 41 (48%). Conclusions Futibatinib treatment resulted in manageable safety, pharmacodynamic activity, and preliminary responses in patients with advanced solid tumors. The results of this phase I dose-escalation trial support 20 mg q.d. futibatinib as the RP2D. Clinical trial registration FOENIX-101 (ClinicalTrials.gov, NCT02052778).

[Lessons Learned] - Palbociclib demonstrated no detectable activity in molecularly unselected and heavily pretreated patients with advanced grade 1/2 pancreatic neuroendocrine tumors. - Predictive biomarkers that improve patient selection should be investigated in future studies of palbociclib., [Background] Palbociclib, a CDK4/6 inhibitor, has shown in vitro activity in pancreatic neuroendocrine tumor (pNET) cell lines. Here we prospectively assessed the activity and safety of palbociclib in monotherapy in metastatic refractory pNETs., [Methods] This was a nonrandomized, open‐label, phase II study of patients with metastatic grade (G)1/2 pNETs recruited from 10 centers in Spain. Palbociclib 125 mg was orally administered once daily for 21 of 28 days until disease progression or unacceptable toxicity., [Results] Twenty‐one patients were included; 52.4% were men, and median age was 57.4 years (range, 37.4–73.4). Patients had previously received a median of three prior lines of systemic therapy (range, 1–10) for advanced disease (somatostatin analogues, 71.4%; sunitinib, 81.0%; everolimus, 47.6%; chemotherapy, 47.6%). Nineteen patients were evaluated for objective response rate (ORR), with a median follow‐up of 12.4 months (range, 7.53–19.33). No objective and confirmed responses were observed (0%); 11 (57.9%) patients had stable disease, and 6 of them lasted more than 6 months; 8 (42.1%) patients had disease progression as best response. Median progression‐free survival (PFS) was 2.6 months (95% confidence interval [CI], 0–14.4) and median overall survival (OS) was 18.7 months (95% CI, 7.4–29.9; Fig. 1). Most frequent toxicities of any grade were asthenia (76.2%), neutropenia (42.9%), diarrhea (33.3%), and nausea (33.3%). Five (23.8%) patients developed G3–4 neutropenia and two (9.5%) patients developed G3–4 thrombocytopenia., [Conclusion] Lack of activity was observed with palbociclib as a single agent in molecularly unselected and heavily pretreated patients with advanced G1/2 pNETs., This work was supported by Pfizer Oncology.

According to the NCI experimental Therapeutics Program (NeXT) definition pharmacodynamic biomarkers (PD) are molecular indicators of the drug’s effect on the target in an organism.

Ramucirumab, a human monoclonal antibody against vascular endothelial growth factor receptor 2 (VEGFR-2), is an antiangiogenic therapy that has been approved in combination with FOLFIRI in second-line treatment of metastatic colorectal cancer (mCRC), after progression on or after therapy with bevacizumab, oxaliplatin, and a fluoropyrimidine. A thorough review of the safety of ramucirumab in this setting and in the context of other antiangiogenic agents is merited. Areas covered: We provide an overview of activity and summarize in detail the overall safety and tolerability profile of ramucirumab in mCRC patients on the basis of a literature review of all published clinical trials in this setting, including both single-agent and combination studies. A focus on adverse events of interest and specific populations is included, as well as a critical comparison with other antiangiogenic therapies. Expert opinion: As an effective agent in pretreated mCRC patients, the toxicity profile of ramucirumab is similar to those of other angiogenesis inhibitors used in the second-line mCRC setting. The next challenge will be to find biomarkers of response and toxicity to antiangiogenic therapies in order to more effectively implement personalized medicine in these patients.

Immunotherapy (IO) has altered the therapeutic landscape for multiple cancers. There are emerging data from retrospective studies on a subset of patients who do not benefit from IO, instead experiencing rapid progression with dramatic acceleration of disease trajectory, termed 'hyperprogressive disease' (HPD). The incidence of HPD ranges from 4% to 29% from the studies reported. Biological basis and mechanisms of HPD are currently being elucidated, with one theory involving the Fc region of antibodies. Another group has shown EGFR and MDM2/MDM4 amplifications in patients with HPD. This phenomenon has polarized oncologists who debate that this could still reflect the natural history of the disease. Thus, prospective studies are urgently needed to confirm the underlying biology, predict patients who are susceptible to HPD, and determine the modality of therapy post progression.

Background Papillary thyroid cancer (PTC) is the most common thyroid carcinoma and exhibits an almost uniformly good prognosis, while anaplastic thyroid cancer (ATC) is less frequent and is one of the most aggressive cancers usually resistant to conventional treatment. Current hypothesis posits that ATC derives from PTC through the progressive acquisition of a discrete number of genomic alterations and implies that the mutational landscape of ATC resembles that of PTC. However, the clinical behaviour of ATC and PTC is radically different. We decided to address the disconnection between the clinical behaviour of ATC and PTC and the proposed model of the progressive development of ATC from PTC. Patients and methods We carried out exome sequencing of DNA from 14 ATC specimens including three cases of concomitant ATC and PTC as well as their corresponding normal DNA from 14 patients. The sequencing results were validated using droplet digital PCR. We carried out immunohistochemistry and immunofluorescence studies of the concomitant ATC and PTC cases. In addition, we integrated our sequencing results with the existing TCGA data. Results Most of the somatic mutations identified in the ATC component differed from the ones in PTC in the cases of concomitant ATC and PTC. The trunks of the phylogenetic trees representing the somatic mutations were short with long branches. In one case of concomitant PTC and ATC specimens, we observed an infiltration of PTC cells within the ATC component. Moreover, we integrated our results with data obtained from TCGA and observed that the most frequent mutations found in ATC presented high cancer cell fraction values and were significantly different from the PTC ones. Conclusion ATC diverge from PTC early in tumour development and both tumour types evolve independently. Our work allows the understanding of the relationship between ATC and PTC facilitating the clinical management of these malignancies.

2524 Background: The phase I GATTO study explored the feasibility, tolerability and preliminary activity of combining gatipotuzumab (GAT), a novel humanized monoclonal antibody binding to the tumor-associated epitope of mucin-1 (TA-MUC1) and an anti-EGFR antibody. Preclinical evidence suggests a complex interaction between TA-MUC1 and EGFR on the cell surface of epithelial tumors driving carcinogenesis processes and synergistic antibody dependent cell cytotoxicity activity with the dual targeting. Methods: Initially the study enrolled in a primary phase (PP) 20 patients with EGFR positive metastatic solid tumors, for whom no standard treatment was available. The first 6 patients were enrolled into a safety run-in phase and the number of dose-limiting toxicities (DLTs) was evaluated, in order to de-escalate the doses if needed. Patients received GAT administered at 1400 mg Q2W in combination with the glyco-optimized anti-EGFR antibody tomuzotuximab (TOM) at 1200 mg Q2W. Due to the risk of infusion related reactions (IRR), the first dose of TOM was reduced to 720 mg split over 2 consecutive days and three cycles of TOM monotherapy were given before start of treatment with GAT. As this regimen was proven safe, no DLT was observed and the initial dose remained unchanged, the study was amended to enroll in an expansion phase (EP) 30 additional patients with refractory colorectal cancer (CRC), non-small cell lung cancer (NSCLC), head and neck and breast cancers. TOM and GAT were given at the same doses and GAT treatment started already one week after the first dose of the anti-EGFR antibody. Additionally investigator had the choice to use a commercial anti-EGFR antibody in place of TOM. Results: By the time of the final analysis in January 2021, 52 refractory patients were enrolled and 50 received at least one dose of both GAT and anti-EGFR antibodies. Panitumumab (PAN) was used in 9 CRC patients. Because of the difference in treatment schedule, results are summarized separately for the 20 and 30 patients in PP and EP. Overall, the combined treatment was well tolerated and no DLT was observed in the whole study, nor related SAE or death. There were no treatment emergent adverse events (TEAEs) leading to dose interruptions or reductions in the PP and 2/30 (6.7%) patients in EP stopped both TOM and GAT. 16 IRRs were reported in 8/20 (40%) PP patients, and 40 IRRs in 10 (33.3%) EP patients. Only one event of chills was severe and only 6 events were related to GAT in the EP, all others to TOM. Other frequent TEAEs were those commonly observed with anti-EGFR treatment such as skin toxicity in 17 (85%) PP and 26 (86.7%) EP patients and hypomagnesemia in 10 (50%) PP and 7 (23.3%) EP patients. Conclusions: Combination of TA-MUC1 and EGFR targeting antibody is safe and feasible. Future studies should test this combination together with chemotherapy Clinical trial information: NCT03360734.

2542 Background: Detection of ctDNA is a promising tool for managing pts in oncology. Most methods require whole-genome sequencing of tumor samples followed by the design of personalized panels for tracking purposes. In this work, we evaluated the prognostic and predictive value of total ctDNA quantification, using shallow whole-genome sequencing (shWGS) exclusively from plasma samples, in a prospective cohort of pts treated with ICIs in early clinical trials. Methods: IchorCNA pipeline was used to quantify ctDNA of shWGS from plasma ctDNA samples of pts treated with ICIs in phase 1 trials, collected at baseline and prior to cycle 2 (prec2). We investigated the association and correlation of ctDNA levels with surrogate markers for tumor burden (LDH levels, summatory of target lesions (TL), liver metastasis) using Spearman and Kruskal-Wallis tests. Kaplan-Meier estimates of overall survival (OS) of pts with baseline detectable ctDNA levels versus undetectable ctDNA were calculated. A multivariate Cox proportional hazards model, including continuous classical prognostic factors (LDH, albumin, hemoglobin, derived Neutrophil-to-Lymphocyte ratio (dNLR), platelets, number of metastases sites, ECOG PS) and ctDNA was performed. An estimate of progression free survival (PFS) of pts with ctDNA increase levels in preC2 versus a non-increase group was evaluated. Results: Since January 2018, 113 pts with no standard-treatment options were included. Median (m) follow up was 14.8 months (mo). Baseline ctDNA levels correlated significantly with baseline TL (R = 0.4, p < 0.001) and LDH levels (R = 0.61, p < 0.001). Pts with liver metastasis had higher levels of ctDNA (11,68 ng/ml) versus pts with no liver disease (2,31 ng/ml) (p < 0,001). In the survival analysis pts with detectable baseline ctDNA (74 pts) had significantly shorter OS compared with pts with undetectable ctDNA (39 pts); median 9.6 m (8.4 – 16.4) and NA m (13.6-NA), respectively (HR = 2.25 [1.18-4.29] p < 0.01). In the multivariate analysis, only ctDNA and albumin levels maintained the impact in OS (HR = 1.03, p < 0.05 and HR = 0.22, p < 0.05, respectively). Pts with early increases in ctDNA had a shorter PFS compared with those with a stabilization or decrease, median 1.9 m (1.6 – 4.0) and 3.0 m (2.6 – 3.8), respectively (HR = 2.19 [1.31-3.67], p < 0.01). Conclusions: Quantification of baseline ctDNA using shWGS is a strong independent prognostic factor. Early dynamic changes of ctDNA could be a useful tool to predict PFS outcomes.

2597 Background: In order to improve the expected response rate (ORR) of less than 10% in cold tumors, several ICI combinations are being evaluated in clinical trials. However, most of these trials don’t require any biomarker and pts are included based solely in histology. We aimed to assess the benefit of ICI combinations in pts with unselected cold tumors included in early CT. Methods: ICI naïve pts with cold tumors treated from 2015 to 2021 with ICI combinations in early CT at VHIO were reviewed. Clinico-pathological data and anti-tumor activity were extracted from a prospective database. ORR was defined as per RECIST v1.1 and clinical benefit rate (CBR) as complete/partial response (CR/PR) + stable disease (SD) for ≥ 4 months (m). Kaplan Meier estimates of progression-free survival (PFS) and overall survival (OS) were calculated and a Cox model according to LIPI (Lung Immune Prognostic Index = baseline LDH and derived neutrophil to lymphocyte ratio) was constructed. Immune-related adverse events (irAE) were classified as per CTCAE v.4.03. Hyperprogressive disease (HPD) was evaluated using RECIST v1.1 (Matos et al, 2020). Results: Out of 97 pts, median age was 62y, 61% had ECOG 0 and 29.8% had LIPI 0 (good prognostic score). Most pts had microsatellite stable (MSS) colorectal cancer (60.8%) or ovarian cancer (14.4%). Regimens included anti-PD1/L1 + another ICI in 69% (most commonly anti-LAG3 [26,8%] and CD40 agonist [20.9%]), anti-PD1/L1 + other molecule in 21.7% (most commonly SHP2 inhibitor [33.3%] and anti p53-HDM2 [28.5%]) and bispecific antibodies in 9.3% (anti-PD1/L1 + anti-LAG3 or CD137 agonist). No patient achieved a response. CBR was 15.3% (11 pts with MSS colorectal cancer, 2 ovarian cancer, 1 olfactory neuroblastoma, 1 paraganglioma). 33 pts (34%) presented irAE, 15 pts (15.5%) had irAE ≥ G2, 4 pts (4.1%) had G3 irAE (dry mouth, hypertransaminasemia, myocarditis and neutrophils count decreased) and 1 patient (1%) had G4 hyperglicemia. 58 pts (59.7%) had progressive disease (PD) as best response, 19 of these pts (32.7%) presented irAE. Overall, 20 pts (20.6%) met definition of HPD, representing 34.4% of pts with PD as best response. Median PFS for overall and CBR population were 1.9 m (CI95% 1.7-2.0) and 5.9 m (5.4-NR), respectively. Median OS for overall population was 7.6 m (5.9-9.5), with a trend for improved OS if LIPI good score vs. others (12.6 m vs. 6.2 m, hazard ratio 1.9, (CI 95% 1.1-3.3), p = 0.02). Among hyperprogressors, median OS was 5.33 m (3.39 - NR) and significantly worse LIPI scores (intermediate [1] or poor [2]) were observed as compared to pts with CBR (75% vs 53.3% p = 0.001). Conclusions: ICI combinations demonstrated very limited activity in pts with unselected cold tumors. However, the risk for irAE and HPD remain substantial. Further drug-biomarker co-development strategies are urgently needed to increase the risk benefit ratio for these pts.

2557 Background: The development of human intratumoral therapy (HIT-IT) has surged as a promising strategy to overcome resistance to checkpoint inhibitors (CPI), promoting a stronger tumor-specific immune response while reducing systemic exposure. A broad variety of agents (i.e: oncolytic viruses, toll-like receptors agonists) administered both in superficial- and deep-seated lesions are being currently tested in clinical trials (CT). Due to the local intervention on tumors, radiological assessment by standard RECIST is challenging and new methods of response that capture and integrate the local and systemic response to HIT-IT are needed. We aimed to evaluate the feasibility and clinical utility of itRECIST (Goldmacher et al., 2020) in patients (pts) treated with HIT-IT in early phase CT. Methods: Retrospective analysis of a cohort of pts with different solid tumor types enrolled in CT including HIT-IT in our institution between August’18 and January’21. Clinical characteristics were collected. Efficacy in target-injected (T-I) and target-non-injected (T-NI) lesions was assessed by objective response rate (ORR) and disease control rate (DCR), as per itRECIST. Overall disease ORR and DCR were assessed per RECIST 1.1/iRECIST. Treatment-related adverse events (TRAEs) were assessed with CTCAE v.5.0. ORR was calculated with Clopper-Pearson method. Survival analysis was made using Kaplan-Meier method. Results: A total of 37 pts were included. Median age was 66 years, 19 pts (51%) were male, all pts had ECOG 0-1. 24 pts (65%) were CPI-naïve. Median previous lines of therapy was 2 (range [r]: 0-11). All pts (100%) received minimum 1 dose of HIT-IT. 6 pts (16%) were treated with monotherapy and 31 pts (84%) in combination with CPI. Median HIT-IT and CPI doses administered were 4 (r: 1-9) and 2 (r: 1-13), respectively. Injected lesions: cutaneous (16.2%), subcutaneous (21.6%), lymph node (32.4%), liver (29.7%). Median size of T-I lesions was 40 mm (r: 19-260). At data cutoff, 32 pts were evaluable. Median follow-up was 14.4 weeks (r: 1.0-81.1). Per RECIST 1.1, overall ORR was 6% (95% CI, 5-7) and DCR was 38% (95% CI, 21-56). Per itRECIST, ORR was 19% (95% CI, 7-36) and DCR was 63% (95% CI, 44-79) in T-I lesions (n = 32), and 10% (95% CI, 22-27) and 48% (95% CI, 29-67) in T-NI lesions (n = 29). Mean decrease in responding T-I and T-NI lesions was -47% (r: -21 to -100) and -41% (r: -26 to -59), respectively. No non-target (NT) lesion was injected. Median progression-free survival was 7.4 weeks (95% CI, 6.6 – 8.2). Median overall survival was 10.0 months (95% CI, 2.3 – 17.7). Incidence of TRAE was 58% (grade 1-2 IT-related pyrexia 43%; grade 3-4, 5%). No treatment-related deaths were recorded. Conclusions: ItRECIST is feasible to implement and adds precision to the radiological assessment of local and distant anti-tumor activity of HIT-IT. No safety issues were detected in our cohort.

2522 Background: The phase I GATTO study explored the feasibility, tolerability and preliminary activity of combining Gatipotuzumab (GAT), a novel humanized monoclonal antibody binding to the tumor-associated epitope of mucin-1 (TA-MUC1), and an anti-EGFR antibody. Preclinical evidence suggests a complex interaction between TA-MUC1 and EGFR on the cell surface of epithelial tumors and synergistic antibody dependent cell cytotoxicity activity with the double targeting. Methods: Initially 20 patients with refractory metastatic disease were treated with GAT administered at 1400 mg Q2W in combination with the glyco-optimized anti-EGFR antibody Tomuzotuximab (TOM) at 1200 mg Q2W. Due to the risk of infusion related reactions, three cycles of TOM were given before start of combined treatment with GAT. After this regimen was proven safe and no DLT was observed, 30 additional patients including colorectal cancer (CRC) already treated with anti-EGFR antibodies, non-small cell lung cancer (NSCLC), head and neck and breast cancers received TOM and GAT administered at the same doses, with GAT treatment starting already one week after the first dose of the anti-EGFR antibody. As allowed in the study expansion, Panitumumab (PAN) was used in place of TOM in 9 CRC patients at investigator’s choice. Results: By the time of the final analysis in January 2021, 52 patients were enrolled, and 50 received at least one dose of both GAT and anti-EGFR antibodies. Safety was overall good and results are reported in a separate abstract. Because of the difference in treatment schedule, activity results of the two parts of the study are summarized separately. There were 2 and 4 RECIST partial responses in the first and second part of the study, all in CRC patients. In the expansion phase, the median Progression Free Survival (PFS) of CRC patients who received TOM (10) and PAN (9) was 1.9 and 5.5 months, respectively. There were 2 responses in each subgroup and the duration of response was 3.8 and 7.2 months in patients receiving TOM and PAN, respectively. The PFS for NSCLC was 5.3 months and 2 heavily pretreated patients achieved a prolonged control of disease of 10.6 and 9.4 months. The trial was accompanied by a comprehensive translational research program for identification of biomarkers, including soluble TA-MUC1 in serum. In the extension phase patients with baseline values above median appeared to have improved PFS and overall survival; this was not the case for patients of the first part of the study who received GAT only after 3 doses of TOM. Conclusions: Combination of TA-MUC1 and EGFR targeting antibody is safe and feasible. Interesting anti-tumor activity was observed in heavily pretreated CRC and NSCLC patients. Levels of soluble TA-MUC1 may have predictive value and potentially be a companion biomarker for further development of the combination Clinical trial information: NCT03360734.

Background & methods: Capecitabine and temozolomide chemotherapy was used in 65 patients with grade 1/2 neuroendocrine tumors (NETs). 46 patients (70.8%) had pancreatic NETs (pNETs). Results: Response rate was 47.7%, with two complete responses (3.1%), 29 partial responses (44.6%) and 27 patients (41.5%) achieved stable disease. Median progression-free survival was 16.1 months (95% CI: 10.7–21.6) and overall survival was 38.3 months (95% CI: 24.6–51.9). Differences in progression-free survival and overall survival between pNETs and non-pNETs were not found. Nine (13.8%) patients experienced grade 3/4 toxicities, mainly thrombocytopenia (10.8%) and neutropenia (7.7%). Conclusion: This is the largest reported series of NETs treated with capecitabine and temozolomide in daily practice and shows that this combination is a promising treatment option for both grade 1/2 pNETs and non-pNETs.