A transcriptomics approach to expand therapeutic options and optimize clinical trials in oncology.

Background: The current model of clinical drug development in oncology displays major limitations due to a high attrition rate in patient enrollment in early phase trials and a high failure rate of drugs in phase III studies. Objective: Integrating transcriptomics for selection of patients has the potential to achieve enhanced speed and efficacy of precision oncology trials for any targeted therapies or immunotherapies. Methods: Relative gene expression level in the metastasis and normal organ-matched tissues from the WINTHER database was used to estimate in silico the potential clinical benefit of specific treatments in a variety of metastatic solid tumors. Results: As example, high mRNA expression in tumor tissue compared to analogous normal tissue of c-MET and its ligand HGF correlated in silico with shorter overall survival (OS; p < 0.0001) and may constitute an independent prognostic marker for outcome of patients with metastatic solid tumors, suggesting a strategy to identify patients most likely to benefit from MET-targeted treatments. The prognostic value of gene expression of several immune therapy targets (PD-L1, CTLA4, TIM3, TIGIT, LAG3, TLR4) was investigated in non-small-cell lung cancers and colorectal cancers (CRCs) and may be useful to optimize the development of their inhibitors, and opening new avenues such as use of anti-TLR4 in treatment of patients with metastatic CRC. Conclusion: This in silico approach is expected to dramatically decrease the attrition of patient enrollment and to simultaneously increase the speed and detection of early signs of efficacy. The model may significantly contribute to lower toxicities. Altogether, our model aims to overcome the limits of current approaches.

Comorbidity between lung cancer and COVID-19 pneumonia: role of immunoregulatory gene transcripts in high ACE2-expressing normal lung.

Background: SARS-CoV-2 (COVID-19) elicits a T-cell antigen-mediated immune response of variable efficacy. To understand this variability, we explored transcriptomic expression of angiotensin-converting enzyme 2 (ACE2, the SARS-CoV-2 receptor) and of immunoregulatory genes in normal lung tissues from patients with non-small cell lung cancer (NSCLC). Methods: This study used the transcriptomic and the clinical data for NSCLC patients generated during the CHEMORES study [n = 123 primary resected (early-stage) NSCLC] and the WINTHER clinical trial (n = 32 metastatic NSCLC). Results: We identified patient subgroups with high and low ACE2 expression (p = 1.55 × 10-19) in normal lung tissue, presumed to be at higher and lower risk, respectively, of developing severe COVID-19 should they become infected. ACE2 transcript expression in normal lung tissues (but not in tumor tissue) of patients with NSCLC was higher in individuals with more advanced disease. High-ACE2 expressors had significantly higher levels of CD8+ cytotoxic T lymphocytes and natural killer cells but with presumably impaired function by high Thymocyte Selection-Associated High Mobility Group Box Protein TOX (TOX) expression. In addition, immune checkpoint-related molecules - PD-L1, CTLA-4, PD-1, and TIGIT - are more highly expressed in normal (but not tumor) lung tissues; these molecules might dampen immune response to either viruses or cancer. Importantly, however, high inducible T-cell co-stimulator (ICOS), which can amplify immune and cytokine reactivity, significantly correlated with high ACE2 expression in univariable analysis of normal lung (but not lung tumor tissue). Conclusions: We report a normal lung immune-tolerant state that may explain a potential comorbidity risk between two diseases - NSCLC and susceptibility to COVID-19 pneumonia. Further, a NSCLC patient subgroup has normal lung tissue expressing high ACE2 and high ICOS transcripts, the latter potentially promoting a hyperimmune response, and possibly leading to severe COVID-19 pulmonary compromise.

A WIN Consortium phase I study exploring avelumab, palbociclib, and axitinib in advanced non-small cell lung cancer.

BACKGROUND: The Worldwide Innovative Network (WIN) Consortium has developed the Simplified Interventional Mapping System (SIMS) to better define the cancer molecular milieu based on genomics/transcriptomics from tumor and analogous normal tissue biopsies. SPRING is the first trial to assess a SIMS-based tri-therapy regimen in advanced non-small cell lung cancer (NSCLC). METHODS: Patients with advanced NSCLC (no EGFR, ALK, or ROS1 alterations; PD-L1 unrestricted; ≤2 prior therapy lines) received avelumab, axitinib, and palbociclib (3 + 3 dose escalation design). RESULTS: Fifteen patients were treated (five centers, four countries): six at each of dose levels 1 (DL1) and DL2; three at DL3. The most common ≥Grade 3 adverse events were neutropenia, hypertension, and fatigue. The recommended Phase II dose (RP2D) was DL1: avelumab 10 mg/kg IV q2weeks, axitinib 3 mg po bid, and palbociclib 75 mg po daily (7 days off/21 days on). Four patients (27%) achieved a partial response (PR) (progression-free survival [PFS]: 14, 24, 25 and 144+ weeks), including two after progression on pembrolizumab. Four patients attained stable disease (SD) that lasted ≥24 weeks: 24, 27, 29, and 64 weeks. At DL1 (RP2D), four of six patients (66%) achieved stable disease (SD) ≥6 months/PR (2 each). Responders included patients with no detectable PD-L1 expression and low tumor mutational burden. CONCLUSIONS: Overall, eight of 15 patients (53%) achieved clinical benefit (SD ≥ 24 weeks/PR) on the avelumab, axitinib, and palbociclib combination. This triplet showed antitumor activity in NSCLC, including in tumors post-pembrolizumab progression, and was active at the RP2D, which was well tolerated. NCT03386929 clinicaltrial.gov.

Digital Display Precision Predictor: the prototype of a global biomarker model to guide treatments with targeted therapy and predict progression-free survival.

The expanding targeted therapy landscape requires combinatorial biomarkers for patient stratification and treatment selection. This requires simultaneous exploration of multiple genes of relevant networks to account for the complexity of mechanisms that govern drug sensitivity and predict clinical outcomes. We present the algorithm, Digital Display Precision Predictor (DDPP), aiming to identify transcriptomic predictors of treatment outcome. For example, 17 and 13 key genes were derived from the literature by their association with MTOR and angiogenesis pathways, respectively, and their expression in tumor versus normal tissues was associated with the progression-free survival (PFS) of patients treated with everolimus or axitinib (respectively) using DDPP. A specific eight-gene set best correlated with PFS in six patients treated with everolimus: AKT2, TSC1, FKB-12, TSC2, RPTOR, RHEB, PIK3CA, and PIK3CB (r = 0.99, p = 5.67E-05). A two-gene set best correlated with PFS in five patients treated with axitinib: KIT and KITLG (r = 0.99, p = 4.68E-04). Leave-one-out experiments demonstrated significant concordance between observed and DDPP-predicted PFS (r = 0.9, p = 0.015) for patients treated with everolimus. Notwithstanding the small cohort and pending further prospective validation, the prototype of DDPP offers the potential to transform patients' treatment selection with a tumor- and treatment-agnostic predictor of outcomes (duration of PFS).

Genomic and transcriptomic profiling expands precision cancer medicine: the WINTHER trial.

Precision medicine focuses on DNA abnormalities, but not all tumors have tractable genomic alterations. The WINTHER trial ( NCT01856296 ) navigated patients to therapy on the basis of fresh biopsy-derived DNA sequencing (arm A; 236 gene panel) or RNA expression (arm B; comparing tumor to normal). The clinical management committee (investigators from five countries) recommended therapies, prioritizing genomic matches; physicians determined the therapy given. Matching scores were calculated post-hoc for each patient, according to drugs received: for DNA, the number of alterations matched divided by the total alteration number; for RNA, expression-matched drug ranks. Overall, 303 patients consented; 107 (35%; 69 in arm A and 38 in arm B) were evaluable for therapy. The median number of previous therapies was three. The most common diagnoses were colon, head and neck, and lung cancers. Among the 107 patients, the rate of stable disease ≥6 months and partial or complete response was 26.2% (arm A: 23.2%; arm B: 31.6% (P = 0.37)). The patient proportion with WINTHER versus previous therapy progression-free survival ratio of >1.5 was 22.4%, which did not meet the pre-specified primary end point. Fewer previous therapies, better performance status and higher matching score correlated with longer progression-free survival (all P < 0.05, multivariate). Our study shows that genomic and transcriptomic profiling are both useful for improving therapy recommendations and patient outcome, and expands personalized cancer treatment.