Predictive and prognostic value of the HLA-G immunocheckpoint in metastatic colorectal cancer treated with FOLFIRI regimen

In recent years, new and innovative immunotherapy drugs entered the market revolutionizing the clinical management of patients suffering of cancer diseases. These strategies aim to reinforce the immune system through the interaction with immune checkpoint inhibitors including human leukocyte antigen-G (HLA-G), due to its immune suppressive properties. The goal is to better understand the natural history of the cancers, the differential outcome in subgroup of patients, response to cancer treatment, develop of toxicity from chemotherapy treatment, all in relation to polymorphisms in immune genes involved in tumor escape mechanisms. The role of HLA-G in the tumorigenesis of metastatic colorectal cancer (mCRC) was here investigated in a retrospective cohort of patients. Despite the improvement of various strategies such as monoclonal antibodies combined with chemotherapy to treat these patients, the 5-year survival rate for mCRC is only slightly over 12%. Moreover, nearly half of mCRC patients are resistant to 5-fluorouracil-based chemotherapies and unresponsive to immunotherapy. Then, novel predictive and prognostic immunobiomarkers are needed to better manage the patients with mCRC. This thesis was divided in three sessions: PART 1, 2 and 3. In PART 1, the predictive and prognostic effect of functional germline polymorphisms in the HLA-G 3’UTR was evaluated in 248 patients with mCRC treated homogeneously with first-line FOLFIRI (Irinotecan (CPT-11) plus 5-Fluorouracil (5-FU) and Leucovorin) regimen. The primary endpoints were OS and TTP. Secondary endpoints were tumor response and develop of severe toxicity. The study endpoints were also evaluated considering particular HLA-G expression level classes, determined on the base of the HLA-G haplotypes. Two “HLA-G secretor models”, were designed for each patient studied, on the base of the reported associations of HLA-G 3’UTR haplotypes with different expression levels of the HLA-G protein: the HLA-G 2 levels (M (Medium/High) level=UTR-1, UTR-3, UTR-4, UTR-6; L (Low) level=UTR-2, UTR-5, UTR-7); and 3 levels (H (High) level=UTR-1; I (Intermediate) level=UTR-2, UTR-3, UTR-4, UTR-6; S (Low) level=UTR-5, UTR-7) secretor models. To exclude a potential confounder factor in the results, we also considered the UGT1A1*28 (rs8175347) polymorphism, that was previously genotyped in the patients included in this study, because of its known impact on Irinotecan toxicity (decreased glucoronidation of the SN-38, the active metabolite of the CPT-11). In PART 2, the soluble HLA-G (sHLA-G) protein was measured in a subgroup of 40 patients with mCRC and studied for the associations with clinical characteristics of the patients and CPT-11 pharmacokinetic parameters; furthermore, the sHLA-G expression was also related to the non-coding regulatory (3’UTR) and coding HLA-G polymorphisms, to search for genotypic/phenotypic relationships. To our knowledge this is the first study to report the genotyping of coding exons for HLA-G mature protein in patients with mCRC. Finally, in PART 3, the intermolecular interactions between HLA-G and CPT-11, that could explain interindividuals variations in the clinical outcome, were for the first time explored. In conclusion, our study suggests a very attractive prognostic and predictive role for the HLA-G molecule in mCRC. Moreover, the evaluation of its levels could improve the management of patients treated with first-line FOLFIRI regimen. The unexplored role of Irinotecan as immunotherapeutic drug that potentially inhibits HLA-G molecule, was for the first time investigated and needs further studies. In conclusion, these findings show novel applications for HLA-G molecule as predictive and prognostic immunobiomarker that could have implications also from a therapeutic point of view, suggesting the develop of a biological drug based on HLA-G structure, that could open new scenarios for the future of the immunotherapy in the fight against cancer.