Gastrointestinal stromal tumors (GISTs) represent 1% of all primary gastrointestinal tumors. Immune surveillance is often overcome by cancer cells due to the activation of immunoregulatory molecules such as programmed death protein (PD-1) and its ligand PD-L1, and butyrophilin sub-family 3A/CD277 receptors (BTN3A). Because several studies demonstrated that tumor PD-1 and PD-L1 expression may have a prominent prognostic function, this investigation aimed to discover if soluble forms of these molecules may be useful in predicting survival of metastatic GIST (mGIST) patients. Through specific ad hoc developed ELISA assays not yet available on the market, the circulating PD-1, PD-L1, BTN3A1, and pan-BTN3As levels were examined in 30 c-KIT exon 11-mutated mGIST patients, prior to imatinib therapy. Using specific thresholds derived by ROC analysis, we found that high baseline levels of sPD-1 (>8.1 ng/mL), sPD-L1 (>0.7 ng/mL), sBTN3A1 (>7.0 ng/mL), and pan-BTN3As (>5.0 ng/mL) were correlated with shorter progression-free survival (PFS) and poor prognosis. Contrariwise, subjects with lower plasma concentrations exhibited a median PFS about 20 months longer than to the earlier. Finally, an additional multivariate analysis revealed that circulating levels of sPD-L1 ≤ 0.7 ng/mL and pan-sBTN3As ≤ 5.0 ng/mL, and the absence of KIT exon 11 deletions or delins at codons 557 and/or 558 were associated with a longer PFS in mGIST patients. Our investigation, for the first time, revealed that evaluating the plasma concentration of some immune checkpoints may help prognosticate survival in mGIST patients, suggesting their potential use as prognostic biomarkers beyond the presence of KIT exon 11 Del or Delins at codons 557/558.

Introduction of checkpoint inhibitors resulted in durable responses and improvements in overall survival in advanced RCC patients, but the treatment efficacy is widely variable, and a considerable number of patients are resistant to PD-1/PD-L1 inhibition. This variability of clinical response makes necessary the discovery of predictive biomarkers for patient selection. Previous findings showed that the epigenetic modifications, including an extensive microRNA-mediated regulation of tumor suppressor genes, are key features of RCC. Based on this biological background, we hypothesized that a miRNA expression profile directly identified in the peripheral lymphocytes of the patients before and after the nivolumab administration could represent a step toward a real-time monitoring of the dynamic changes during cancer evolution and treatment. Interestingly, we found a specific subset of miRNAs, called “lymphocyte miRNA signature”, specifically induced in long-responder patients (CR, PR, or SD to nivolumab >18 months). Focusing on the clinical translational potential of miRNAs in controlling the expression of immune checkpoints, we identified the association between the plasma levels of soluble PD-1/PD-L1 and expression of some lymphocyte miRNAs. These findings could help the development of novel dynamic predictive biomarkers urgently needed to predict the potential response to immunotherapy and to guide clinical decision-making in RCC patients.

e21525 Background: Merkel cell carcinoma (MCC) is a rare and aggressive skin cancer, associated with a worse prognosis. The link between MCC and immune suppression is well demonstrated. The population of patients with MCC is frequently elderly and frail, making it essential to determine whether the results of clinical trial can be replicated in a real-world setting. Despite the introduction of immune-checkpoint inhibitors (ICIs) has provided great benefit for some patients with advanced MCC (aMCC), it remains a subsets of patients who are refractory to ICIs or develop acquired resistance over time. Thus, there is a clinical need for predictive factors of ICI response. Methods: Twenty patients with aMCC treated with avelumab were included. The treatment was administered as I or II line. Clinical-pathological characteristics, Body Mass Index (BMI), and response to avelumab were analyzed from a MCC System database prospectively collected. An explorative analysis was performed, for available samples, to investigate: i) the plasma levels of soluble PD-1 (sPD-1), and PD-L1 (sPD-L1) collected at baseline, measured using homemade ELISA assays not yet commercially available, and designed according to investigator specifications; ii) IHC for PD-L1 in tumor samples; iii) the presence/absence/class (brisk vs no-brisk) of tumor-infiltrating lymphocytes (TILs) in tumor samples. The primary outcome investigated was the Time to Treatment Failure (TTF). Results: From February 2019 to January 2022, twenty (20) patients were included in the study. The median age was 74 (range 56-83); 10 patients were men (50%) and 10 were women (50%). Seventeen (17) patients (85%) were treated with avelumab as I line, and 3 patients (15%) as II line. The overall response rate was 65% (70.6% in I line patients). One (1) patient (5%) had a complete response (CR), 13 patients (65%) partial response (RP), 4 patients (20%) stable disease (SD), and 2 patients (10%) had a progression disease (PD). Overall median TTF was 22 months (95% confidence interval [CI]: -13.0-30.9). At the time of data analyses, a total of 9 events (progression or death) occurred (45%). Notably, a BMI ≥ 30 was significantly associated with longer TTF (p = 0.004) and objective response rate (p = 0.01). In the explorative biomarker analysis, preliminary data on 6 tumor and plasma samples, showed that plasma sPD-1 > 3.8 ng/ml, and the presence of PD-L1 and brisk TILs on tumor samples, were associated to longer TTF. Conclusions: These finding highlight the complex immune-metabolic interplay in the immunotherapy response. These data extends the previous finding on “obesity paradox” and the role of BMI as predictive factors of ICIs. The data on biomarker analysis warrants further prospective validation.

Gastrointestinal Stromal Tumors (GISTs) represent a paradigmatic model of oncogene addiction. Despite the well-known impact of the mutational status on clinical outcomes, we need to expand our knowledge to other factors that influence behavior heterogeneity in GIST patients. A growing body of studies has revealed that the tumor microenvironment (TME), mostly populated by tumor-associated macrophages (TAMs) and lymphocytes (TILs), and stromal differentiation (SD) have a significant impact on prognosis and response to treatment. Interestingly, even though the current knowledge of the role of immune response in this setting is still limited, recent pre-clinical and clinical data have highlighted the relevance of the TME in GISTs, with possible implications for clinical practice in the near future. Moreover, the expression of immune checkpoints, such as PD-L1, PD-1, and CTLA-4, and their relationship to the clinical phenotype in GIST are emerging as potential prognostic biomarkers. Looking forward, these variables related to the underlying tumoral microenvironment in GIST, though limited to still-ongoing trials, might lead to the potential use of immunotherapy, alone or in combination with targeted therapy, in advanced TKI-refractory GISTs. This review aims to deepen understanding of the potential link between mutational status and the immune microenvironment in GIST.

The volume estimation of retroperitoneal sarcoma (RPS) is often difficult due to its huge dimensions and irregular shape; thus, it often requires manual segmentation, which is time-consuming and operator-dependent. This study aimed to evaluate two fully automated deep learning networks (ENet and ERFNet) for RPS segmentation. This retrospective study included 20 patients with RPS who received an abdominal computed tomography (CT) examination. Forty-nine CT examinations, with a total of 72 lesions, were included. Manual segmentation was performed by two radiologists in consensus, and automatic segmentation was performed using ENet and ERFNet. Significant differences between manual and automatic segmentation were tested using the analysis of variance (ANOVA). A set of performance indicators for the shape comparison (namely sensitivity), positive predictive value (PPV), dice similarity coefficient (DSC), volume overlap error (VOE), and volumetric differences (VD) were calculated. There were no significant differences found between the RPS volumes obtained using manual segmentation and ENet (p-value = 0.935), manual segmentation and ERFNet (p-value = 0.544), or ENet and ERFNet (p-value = 0.119). The sensitivity, PPV, DSC, VOE, and VD for ENet and ERFNet were 91.54% and 72.21%, 89.85% and 87.00%, 90.52% and 74.85%, 16.87% and 36.85%, and 2.11% and -14.80%, respectively. By using a dedicated GPU, ENet took around 15 s for segmentation versus 13 s for ERFNet. In the case of CPU, ENet took around 2 min versus 1 min for ERFNet. The manual approach required approximately one hour per segmentation. In conclusion, fully automatic deep learning networks are reliable methods for RPS volume assessment. ENet performs better than ERFNet for automatic segmentation, though it requires more time.