Epithelial ovarian cancers (EOC) associated with germline or somatic BRCA pathogenetic variants have a significantly higher rate of TP53aberrations. The majority of TP53 mutations are detectable by immunohistochemistry and several studies demonstrated that an abnormal p53 pattern characterized high-grade EOCs. An abnormal p53 immunohistochemical staining in fallopian tube (serous tubal intraepithelial carcinoma (STIC) and "p53 signature" is considered as a precancerous lesion of high-grade EOCs and it is often found in fallopian tube tissues of BRCA germline mutated patients suggesting that STIC is an early lesion and the TP53 mutation is an early driver event of BRCA mutated high-grade EOCs. No relevant data are present in literature about the involvement of p53 abnormal pattern in EOC carcinogenesis of patients negative for germline BRCA variants. We describe TP53 mutation results in relationship to the immunohistochemical pattern of p53 expression in a series of EOCs negative for BRCA1 and BRCA2 germline mutations. In addition, we also investigated STIC presence and "p53 signature" in fallopian tube sampling of these EOCs. Our results demonstrate that TP53 alterations are frequent and early events in sporadic EOCs including also low-grade carcinomas. Also in this series, STIC is associated with an abnormal p53 pattern in fallopian tubes of high-grade EOCs. In summary, TP53 aberrations are the most frequent and early molecular events in EOC carcinogenesis independently from BRCA mutation status.

BACKGROUND:The clinical utility of germline BRCA1 and BRCA2 testing is well established in patients with family history suggestive for hereditary breast and ovarian cancer syndrome. Recently, germline PALB2 pathogenic variants were also associated with an increased risk of breast and other cancers and, in the Italian population, it has been described in few studies without a systematic germline analysis of BRCA1, BRCA2 and PALB2. OBJECTIVES AND METHODS:In this study, we described ASST Sette Laghi cancer genetic counselling services' experience in the analysis of 402 patients with suspected breast and ovarian cancer syndrome, by using BRCA1, BRCA2 and PALB2 germline genetic test. RESULTS:The frequency of PALB2 pathogenic variants was 1.2% compared to 3.5% and 3.2% for BRCA1 and BRCA2, respectively, whereas class 3 variants were detected in 0.3% and 0.5% of the BRCA1 and BRCA2 investigated patients, respectively. PALB2 pathogenic variants were identified in patients with a strong family history for breast cancer. Moreover, PALB2 variants were significantly associated with a younger age of breast cancer onset (mean age, 40.25 years) compared to wild-type patients (mean age 51.2 years, p-value = 0.0331). Similar to BRCA-associated breast cancer, the majority of PALB2 breast cancers were identified at an advanced clinical stage. Pedigree analysis revealed a family history of breast and ovarian cancer syndrome in all PALB2 pathogenic variants carriers (early breast cancer onset, bilateral breast cancer and ovarian cancer). CONCLUSION:In conclusion, the germline analysis of BRCA1, BRCA2 and PALB2 should be included in breast cancer clinical practice as a not negligible number of PALB2 carriers could be identified and referred to specific surveillance protocols.

We analyzed the clinicopathological, cytogenetic, and molecular features of 18 primary cutaneous diffuse large B-cell lymphomas (PCDLBCLs) and 15 DLBCLs secondarily localized to the skin (SCDLBCLs), highlighting biological similarities and differences between the 2 groups. PCDLBCLs were subclassified after histopathological review as PCDLBCL-leg type (PCDLBCL-LT, 10 cases) and the PCDLBCL-not otherwise specified (PCDLBCL-NOS, 8 cases). Immunohistochemistry for Hans' algorithm markers, BCL2, and MYC was performed. The molecular study included the determination of the cell of origin (COO) by Lymph2Cx assay on NanoString platform, FISH analysis of IgH, BCL2, BCL6, and MYC genes, as well as the mutation analysis of MYD88 gene. In immunohistochemistry analysis, BCL2 and MYC hyperexpression was more frequent in LT than in NOS cases and, according to Hans' algorithm, PCDLBCL-LTs were mostly of the non-GC type (8/10), whereas in PCDLBCL-NOS, the GC type prevailed (6/8). The determination of COO using Lymph2Cx supported and further confirmed these results. In FISH analysis, all but one LT cases versus 5 of 8 PCDLBCL-NOS showed at least one gene rearrangement among IgH, BCL2, MYC, or BCL6. In addition, MYD88 mutations were more frequently present in LT than in NOS subtypes. Interestingly, MYD88-mutated patients were older, with a non-GC phenotype and had worse OS, compared to MYD88 WT cases. Overall, SCDLBCL did not show, at the genetic and expression level, different profiles than PCDLBCL, even if they bear a significantly worse prognosis. At survival analysis, the most important prognostic factors in patients with PCDLBCL were age and MYD88 mutation, whereas relapse and high Ki-67 expression were relevant in patients with SCDLBCL. Our study comprehensively analyzed the clinicopathological and molecular features of PCDLBCL-LT, PCDLBCL-NOS, and SCDLBCL, underlining the differences among them and the importance of properly identifying these entities at the time of diagnosis.

Epithelial ovarian cancers (EOCs) harboring germline or somatic pathogenic variants in BRCA1 and BRCA2 genes show sensitivity to poly(ADP-ribose) polymerase inhibition. It has been suggested that BRCA1 promoter methylation is perhaps a better determinant of therapy response, because of its intrinsic dynamic feature, with respect to genomic scars or gene mutation. Conflicting evidence was reported so far, and the lack of a validated assay to measure promoter methylation was considered a main confounding factor in data interpretation. To contribute to the validation process of a pyrosequencing assay for BRCA1 promoter methylation, 109 EOCs from two Italian centers were reciprocally blindly investigated. By comparing two different pyrosequencing assays, addressing a partially overlapping region of BRCA1 promoter, an almost complete concordance of results was obtained. Moreover, the clinical relevance of this approach was also supported by the finding of BRCA1 transcript down-regulation in BRCA1-methylated EOCs. These findings could lead to the development of a simple and cheap pyrosequencing assay for diagnostics, easily applicable to formalin-fixed, paraffin-embedded tissues. This technique may be implemented in routine clinical practice in the near future to identify EOCs sensitive to poly(ADP-ribose) polymerase inhibitor therapy, thus increasing the subset of women affected by EOCs who could benefit from such treatment.

Epithelial ovarian cancers (EOCs) harboring germline or somatic pathogenic variants in BRCA1 and BRCA2 genes show sensitivity to poly(ADP-ribose) polymerase inhibition. It has been suggested that BRCA1 promoter methylation is perhaps a better determinant of therapy response, because of its intrinsic dynamic feature, with respect to genomic scars or gene mutation. Conflicting evidence was reported so far, and the lack of a validated assay to measure promoter methylation was considered a main confounding factor in data interpretation. To contribute to the validation process of a pyrosequencing assay for BRCA1 promoter methylation, 109 EOCs from two Italian centers were reciprocally blindly investigated. By comparing two different pyrosequencing assays, addressing a partially overlapping region of BRCA1 promoter, an almost complete concordance of results was obtained. Moreover, the clinical relevance of this approach was also supported by the finding of BRCA1 transcript down-regulation in BRCA1-methylated EOCs. These findings could lead to the development of a simple and cheap pyrosequencing assay for diagnostics, easily applicable to formalin-fixed, paraffin-embedded tissues. This technique may be implemented in routine clinical practice in the near future to identify EOCs sensitive to poly(ADP-ribose) polymerase inhibitor therapy, thus increasing the subset of women affected by EOCs who could benefit from such treatment.

(1) Background: MLH1 hypermethylation is an epigenetic alteration in the tumorigenesis of colorectal cancer (CRC) and endometrial cancer (EC), causing gene silencing, and, as a consequence, microsatellite instability. Commonly, MLH1 hypermethylation is considered a somatic and sporadic event in cancer, and its detection is recognized as a useful tool to distinguish sporadic from inherited conditions (such as, Lynch syndrome (LS)). However, MLH1 hypermethylation has been described in rare cases of CRC and EC in LS patients. (2) Methods: A total of 61 cancers (31 CRCs, 27 ECs, 2 ovarian cancers, and 1 stomach cancer) from 56 patients referred to cancer genetic counselling were selected for loss of MLH1 protein expression and microsatellite instability. All cases were investigated for MLH1 promoter methylation and MLH1/PMS2 germline variants. (3) Results: Somatic MLH1 promoter hypermethylation was identified in 16.7% of CRC and in 40% of EC carriers of MLH1 germline pathogenic variants. In two families, primary and secondary MLH1 epimutations were demonstrated. (4) Conclusions: MLH1 hypermethylation should not be exclusively considered as a sporadic cancer mechanism, as a non-negligible number of LS-related cancers are MLH1 hypermethylated. Current flow charts for universal LS screening, which include MLH1 methylation, should be applied, paying attention to a patient’s family and personal history.

Introduction:BRCA1 methylated (BRCA1met) epithelial ovarian cancer (EOC) is a recently defined and not well-investigated subset of neoplasms. To date, no studies have focused on the transcriptional profiles of BRCA1met cases, and, as a matter of fact, we still do not know if this subset of EOCs is similar, and to what extent, to BRCA1 mutated (BRCA1mut) cases.Methods:We compared a group of 17 BRCA1met cases against 10 BRCA1mut cases using a subset of carefully selected 17 BRCAwt EOCs as a control group.Results:First, BRCA1met cases showed a downregulation of the relative transcript, while this association was not observed for BRCA1mut EOCs. The BRCA1met group exhibited a general upregulation of homologous recombination (HR)-related genes, as well as BRCA1mut. Overall, BRCA1met had a different gene expression profile, characterized by diffuse downregulation, whereas BRCA1mut showed a general upregulation (p < 0.0001). Both BRCA1-defective groups showed a slightly activated immune response mediated by interferon (IFN) gamma pathways.Discussion:In conclusion, even if the expression profile of many genes related to DNA damage and repair system is shared between BRCA1mut and BRCA1met EOCs supporting that BRCA1met EOCs may benefit from PARPi therapies, our data demonstrate that BRCA1mut and BRCA1met EOCs show different expression profiles, suggesting a different mechanism of carcinogenesis that can be reflected in different responses to therapies and disease recovery.

BCL2 rearrangement is reported to be an early pathogenetic event in follicular lymphoma (FL) and it is considered as a reliable marker in the follow up of the disease. We aimed to investigate the frequency of BCL2 rearrangement in FLs from northwestern Italy, to evaluate their clinicopathological features, and to investigate alternative genetic aberrations in BCL2-negative FLs. We collected a series of 76 consecutive FLs diagnosed between 2013 and 2016. All lymphomas underwent histopathological review. Interphasic fluorescent in situ hybridization (FISH) was performed with break apart probes targeting BCL2, IGH, BCL6 and MYC on paraffin embedded (PE) and fresh frozen (FF) specimens. 1p36 region and p53 locus in BLC2-negative cases were investigated using dual color probes. Karyotype analysis was available in a subset of cases. BCL2 rearrangements were detected in 39 cases (51,3%). Of the remaining 37, 6 showed IGH rearrangement, and were further tested: 1 showed variant BCL2 translocation, 1 had BCL6 rearrangement, and the other 4 were negative for further gene rearrangements. FISH on FF specimens detected small BCL2+ clones in cases otherwise categorized as BCL2-. 1p36 and p53 deletion were observed in 1 and 8 BCL2- FLs, respectively. Karyotype analysis documented 3q, 1p and BCL6 alternative abnormalities in 3 cases. In conclusion, BCL2 rearrangement is not a constant finding in FL, its frequency being probably affected by geographical factors. Thus, it should not be considered as a reliable molecular marker in the follow up of the disease, unless it is found to be present at the initial diagnosis of FL. Alternative genetic aberrations exist in BCL2-negative cases.

Background The classification of PCDLBCL and the exact distinction between NOS and LT are still a matter of debate. To contribute to this discussion we analyzed the complete clinico-pathological, cytogenetic and molecular features of a series of PCDLBCL and, for comparison, of DLBCL secondarily localized to the skin, diagnosed and treated in 4 Italian centers. Here, we highlight biologic similarities and differences between primary and secondary cases, particularly focusing on the cytogenetic and molecular features. Patients and Method We analyzed clinical data and skin biopsies of pts diagnosed as having PCDLBCL when they had no previous history of DLBCL and other lesions were excluded by clinical and radiological examination. All other pts were considered as having SCDLBCL. During a session at a multi-head microscope, the pathologists reviewed all slides. By Hans algorithm, cases were classified as GC and non-GC. Based on the histopathological and IHC features and clinical data, PCDLBCL were segregated into PCDLBCL-LT and PCDLBCL-NOS (Lucioni M., Cancer Med, 2016). FISH analysis was performed and interpreted at the Laboratory of cytogenetic of Varese University hospital as previously described (Tibiletti M.G., Cancer Genet Cytogenet, 2004). On DNA and RNA from FFPE, we performed the analysis of MYD88 mutations by NGS and COO classification by the Lymph2Cx assay that subdivides DLBCL into ABC, GCB and Unclassified subsets. Results Clinical characteristics of the 33 CDLBCL pts are summarized in Tab 1. Figure 1 summarizes the clinical data and pathological, immunophenotypic, cytogenetic and molecular results of 18 PCDLBCL and 15 SCDLBCL pts. PCDLBCL were classified as LT and NOS in 56% and 44%, respectively. BCL2, MYC and DE phenotype were less expressed in NOS pts,50%, 29% and 14%, respectively than in LT cases ,100%, 50% and 50%, respectively. The results of the interesting comparison between PCDLBCL-NOS GCB with the remaining PCLBCL are reported in Tab1. Lymph2Cx subdivided the 28 CDLBCL in 17 ABC, 7 GCB and 4 Unclassified subtypes. Among PCDLBCL-LT pts, we did not find any GCB case, while it represented the most frequent type among PCDLBCL-NOS pts. Classification by COO and Hans algorithm was concordant in 64 % of pts. FISH analysis showed that 23 of the 33 pts had at least one gene rearrangement, including 13/18 pts with PCDLBCL and 10/15 pts with SCDLBCL. All cases had IGH rearrangement. MYC rearrangement was found in 5 pts with PCDLBCL and in 6 with SCDLBCL. BLC6 was rearranged in 4 pts with PCDLBCL and in 3 with SCDLBCL, while BCL2 was rearranged in 1 patient with PCDLBCL and in 2 with SCDLBCL. Finally, we observed that 8/9 pts with PCDLBCL-LT had at least one driver gene rearranged, while only 5/9 PCDLBCL-NOS were rearranged.MYD88 mutations were more frequent in PCDLBCL- LT than in PCDLBCL-NOS. Among SCDLBCL cases, MYD88 mutation was found in 47% of pts with the non-GC subtype. Significantly, none of the GC pts resulted positive. MYD88 mutated pts were older and with a non-GCB phenotype, compared to MYD88 WT pts. MYD88 WT CDLBCL pts had a significant better OS compared to the pts with L265P mutation. Among the PCDLBCL pts, in terms of both OS and PFS the most relevant clinical prognostic factors were age and MYD88 mutation. Among the SCDLBCL, only relapse and high expression of Ki67 are associated with shorter OS (Tab1). Discussion Our study emphasizes the clinico-pathological, cytogenetic and molecular similarities and differences between PCDLBCL and SCDLBCL, underlying the importance of properly staging CDLBCL pts at time of diagnosis . In the PCDLBCL setting, we confirmed the independent pathological entity of the NOS category highlighting the better prognostic outcome of this subtype. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Ovarian cancer (OC) is the fifth most common type of cancer in women and the fourth most common cause of cancer death in women. Identification of pathogenic variants in OC tissues has an important clinical significance for therapeutic and prevention purposes. This study aims to evaluate the mutational profile of a patient cohort, negative for BRCA1/2 germinal variants and Mismatch Repair defects, using next-generation sequencing (NGS) approach on DNA from formalin-fixed paraffin-embedded samples. We used a custom NGS panel, targeting 34 cancer-related genes, mainly of the BRCA and PARP pathways, and analyzed NGS data to identify somatic and germline variants in Italian patients affected by primary epithelial ovarian cancer. We analyzed 75 epithelial ovarian cancer tissues and identified 54 pathogenic variants and 56 variants of unknown significance. TP53 was characterized by the highest mutational rate, occurring in 55% of tested epithelial ovarian cancers (EOCs). Interestingly, a subset of 8 EOCs showed pathogenic variants of homologous recombination pathway, which could be sensitive to PARP-inhibitor therapies. Germline analysis of actionable genes revealed 4 patients carrier of pathogenic germline variants respectively of RAD51C (2 patients), RAD51D, and PALB2. Molecular profiling of EOCs using our custom NGS panel has enabled the detection of both somatic and germline variants, allowing the selection of patients suitable for targeted therapies, and the identification of high-risk OC families that can benefit from genetic counseling and testing.

Pancreatic cancer has a high morbidity and mortality with the majority being PC ductal adenocarcinomas (PDAC). Whole genome sequencing provides a wide description of genomic events involved in pancreatic carcinogenesis and identifies putative biomarkers for new therapeutic approaches. However, currently, there are no approved treatments targeting driver mutations in PDAC that could produce clinical benefit for PDAC patients. A proportion of 5–10% of PDAC have a hereditary origin involving germline variants of homologous recombination genes, such as Mismatch Repair (MMR), STK11 and CDKN2A genes. Very recently, BRCA genes have been demonstrated as a useful biomarker for PARP-inhibitor (PARPi) treatments. In this study, a series of 21 FFPE PDACs were analyzed using OncoPan®, a strategic next-generation sequencing (NGS) panel of 37 genes, useful for identification of therapeutic targets and inherited cancer syndromes. Interestingly, this approach, successful also on minute pancreatic specimens, identified biomarkers for personalized therapy in five PDAC patients, including two cases with HER2 amplification and three cases with mutations in HR genes (BRCA1, BRCA2 and FANCM) and potentially eligible to PARPi therapy. Molecular analysis on normal tissue identified one PDAC patient as a carrier of a germline BRCA1 pathogenetic variant and, noteworthy, this patient was a member of a family affected by inherited breast and ovarian cancer conditions. This study demonstrates that the OncoPan® NGS-based panel constitutes an efficient methodology for the molecular profiling of PDAC, suitable for identifying molecular markers both for therapy and risk assessment. Our data demonstrate the feasibility and utility of these NGS analysis in the routine setting of PDAC molecular characterization.

Background Endometrial carcinoma represents a sentinel cancer for Lynch syndrome (LS) identification. It is crucial to highlight how other types of tumors can arise in the gynecological tract acting as sentinel tumors in LS patients. Up to now, no established LS patient management strategy has incorporated the presence of these additional candidate sentinel tumors to improve the prevention and management of LS tumors. Methods In order to investigate the involvement of the most frequent gynecological cancers in gynecological cancers, we studied different subsets of gynecological cancers using both somatic approaches, including mismatch repair (MMR) gene immunohistochemical expression, microsatellite instability, and germline analyses ofMSH2, MSH6, MLH1, PMS2 and EPCAM genes. A total of 261 patients referring to the Cancer Genetic Counselling Service of our institution were included in the study. In detail, our series was composed of 131 patients affected by uterus cancers including endometrial, isthmus and non-HPV endocervical carcinomas, 113 patients affected by ovarian cancers and 17 patients affected by synchronous endometrial/ovarian carcinomas (SEOC). In addition, we studied 115 cases of endometrial cancers identified by 2 years of universal testing (endometrial cancers/UTs) using IHC analysis of four MMR proteins. Results and conclusions The incidence of MMR defective gynecological cancers ranged from 7.1 to 47.1% depending on cancer site and selection. LS patients carriers of pathogenetic MMR variants were identified in 19.8% of uterus cancers, 35.3% of SEOC, 4.4% of ovarian cancers. In addition, pathogenetic MMR variants were identified in 4.3% of endometrial cancers/universal testing investigated with universal screening. In conclusion, gynecological cancers are heavily involved in LS and our study shows that MMR screening using immunohistochemical pattern and MSI analysis of endometrial and ovarian cancers as well as of rare entities such as non-HPV related endocervical cancers and synchronous endometrial and ovarian cancers are sentinels for LS. Tumor testing approach improves early identification of MMR defective gynecological cancers and this is an effective strategy to detect high-risk patients and to offer them and their relatives personalized cancer prevention.

Homologous Recombination Deficiency (HRD) is a frequent feature of high-grade epithelial ovarian carcinoma (EOC), associated with sensitivity to PARP-inhibitors (PARPi). The best characterized causes of HRD in EOCs are germline or somatic mutations in BRCA1 and BRCA2 genes. Although promoter methylation is a well-known mechanism of gene transcriptional repression, few data have been published about BRCA gene methylation in EOCs. In this retrospective study, we quantitatively analyzed by pyrosequencing a selected series of 90 formalin-fixed (FFPE) primary EOCs without BRCA germline mutations. We identified 20/88 (22.7%) EOCs showing BRCA promoter methylation, including 17/88 (19.3%) in BRCA1 and 4/86 (4.6%) in BRCA2 promoters, one of which showing concomitant BRCA1 methylation. Mean methylation levels were 49.6% and 45.8% for BRCA1 and BRCA2, respectively, with methylation levels ≥50% in 10/20 methylated EOCs. Constitutive BRCA methylation was excluded by testing blood-derived DNA. In conclusion, pyrosequencing methylation analysis of BRCA genes is a robust, quantitative and sensitive assay applicable to FFPE samples. Remarkably, a considerable subset of germline BRCA-negative EOCs showed somatic methylation and, likely, HRD. A subpopulation of women with BRCA methylation, even without BRCA mutations, could potentially benefit from PARP-inhibitors; further clinical studies are needed to clarify the predictive role of somatic BRCA methylation of PARP-therapy response.