L. Belvederesi

Insulin-like growth factor (IGF) 1 and 2 help to predict disease outcome in GIST patients

BACKGROUND The expression of the insulin-like growth factor (IGF) system has never been studied in gastrointestinal stromal tumors (GISTs). PATIENTS AND METHODS We studied the immunohistochemical expression of IGF1 receptor (IGFR-I), IGF1 and IGF2 in 94 samples of GISTs. IGF1 and IGF2 expression was scored in three classes: negative (N), moderate (M) and strong (S), according to staining intensity and extent. RESULTS IGFR-I was overexpressed in all cases. IGF1 and IGF2 expression was absent in 25 and 48 cases, moderate in 29 and 16 cases and strong in 40 and 30 cases, respectively. Strong IGF1 expression significantly correlated with higher mitotic index (P = 0.0001), larger (P = 0.01), higher risk (P = 0.0002), metastatic (P = 0.0001) and relapsed (P = 0.04) GISTs. Strong IGF2 expression correlated with higher mitotic index (P = 0.05) and higher risk GISTs (P = 0.001). The Kaplan-Meier analysis (N versus M versus S) showed a significant worsening of the disease-free survival (DFS) with the increase of IGF1 (P = 0.02) and IGF2 (P = 0.02) expression. In the subgroup of patients with operated high-risk GISTs, there was a better trend in DFS for patients affected by GISTs with negative IGF1 and IGF2. CONCLUSIONS The expression of IGF1 and IGF2 seems to predict relapse in GIST patients.

Pre-treatment lactate dehydrogenase levels as predictor of efficacy of first-line bevacizumab-based therapy in metastatic colorectal cancer patients

Background:Lactate dehydrogenase (LDH) represents a predictive factor in colorectal cancer patients treated with the angiogenesis inhibitor PTK/ZK. We explored the role of pre-treatment LDH serum levels in colorectal cancer patients receiving first-line bevacizumab.Methods:Metastatic colorectal cancer treated with first-line bevacizumab was eligible. A control group including all consecutive patients treated with chemotherapy alone was also considered. Pre-treatment LDH serum levels were collected for all cases.Results:Median progression-free survival (PFS) in the control group for patients with high and low LDH levels was 4.2 and 8 months, respectively (P=0.0003). Median overall survival (OS) was 19.6 and 34.9 months for patients with high and low LDH levels, respectively (P=0.0014). In the bevacizumab group, partial responses were seen in 14 (58%) high-LDH and 8 (14%) low-LDH patients (P=0.0243), respectively, median PFS was 7.3 and 8.5 months, respectively (P=0.2), and median OS was 22 and 26.6 months, respectively (P=0.7).Conclusion:High LDH levels correlated with worse prognosis. Bevacizumab seemed capable of improving clinical outcome in this specific group of patients who usually present with an adverse natural history. The improved response rate also suggests a role for LDH as a predictive marker.

The Role of LDH Serum Levels in Predicting Global Outcome in HCC Patients Undergoing TACE: Implications for Clinical Management

In many tumor types serum lactate dehydrogenase (LDH) levels is an indirect marker of tumor hypoxia, neo-angiogenesis and worse prognosis. However data about hepatocellular carcinoma (HCC) are lacking in the clinical setting of patients undergoing transarterial-chemoembolization (TACE) in whom hypoxia and neo-angiogenesis may represent a molecular key to treatment failure. Aim of our analysis was to evaluate the role of LDH pre-treatment levels in determining clinical outcome for patients with HCC receiving TACE. One hundred and fourteen patients were available for our analysis. For all patients LDH values were collected within one month before the procedure. We divided our patients into two groups, according to LDH serum concentration registered before TACE (first: LDH≤450 U/l 84 patients; second: LDH>450 U/l 30 patients). Patients were classified according to the variation in LDH serum levels pre- and post-treatment (increased: 62 patients vs. decreased 52 patients). No statistically significant differences were found between the groups for all clinical characteristics analyzed (gender, median age, performance status ECOG, staging systems). In patients with LDH values below 450 U/l median time to progression (TTP) was 16.3 months, whereas it was of 10.1 months in patients above the cut-off (p = 0.0085). Accordingly median overall survival (OS) was 22.4 months and 11.7 months (p = 0.0049). In patients with decreased LDH values after treatment median TTP was 12.4 months, and median OS was 22.1 months, whereas TTP was 9.1 months and OS was 9.5 in patients with increased LDH levels (TTP: p = 0.0087; OS: p<0.0001). In our experience, LDH seemed able to predict clinical outcome for HCC patients undergoing TACE. Given the correlation between LDH levels and tumor angiogenesis we can speculate that patients with high LDH pretreatment levels may be optimal candidates for clinical trial exploring a multimodality treatment approach with TACE and anti-VEGF inhibitors in order to improve TTP and OS.

Assessing the pathogenicity of MLH1 missense mutations in patients with suspected hereditary nonpolyposis colorectal cancer: correlation with clinical, genetic and functional features.

Assessing the pathogenicity of missense mutations of MLH1 and MSH2 is critical to counsel patients with suspected hereditary nonpolyposis colorectal cancer (HNPCC). Approximately 32% of all MLH1 mutations and 18% of MSH2 mutations are missense variants which often have an uncertain genetic significance. To assess the pathogenicity of four MLH1 missense mutations which were found in five patients with suspected HNPCC, P648S (CCC → TCC), L559R (CTG → CGG), K618A (AAG → GCG), Y646C (TAT → TGT), we studied their ability to disrupt MLH1 protein function and their relationship with all those clinical, genetic and pathological features which are typical of this syndrome. Our results indicated that the P648S and L559R mutations were probably pathogenic because they disrupted MLH1 protein interaction with its partner PMS2 in vitro and abolished MLH1 expression in HCT116 cells. In addition these variants were associated with features often found in HNPCC patients: in particular high microsatellite instability, occurrence of high grade tumours and, in one case, strong family history. The pathogenicity of the K618A and Y646C mutations was questionable as their correlation with features typical of HNPCC was low and the outcome of the functional analysis was ambiguous. These observations suggested that a clinically usable assessment of the pathogenicity of MLH missense variants can be achieved through the analysis of multiple mutation characteristics among which loss of protein function, occurrence of microsatellite instability and family history seemed to have a predominant role.

Prognostic impact of mismatch repair genes germline defects in colorectal cancer patients: are all mutations equal?

Background Lynch syndrome (LS) is the most common hereditary colorectal cancer (CRC) syndrome, caused by germline mutations in MisMatch Repair (MMR) genes, particularly in MLH1, MSH2 and MSH6. Patients with LS seem to have a more favourable prognosis than those with sporadic CRC, although the prognostic impact of different mutation types is unknown. Aim of our study is to compare survival outcomes of different types of MMR mutations in patients with LS-related CRC. Methods 302 CRC patients were prospectively selected on the basis of Amsterdam or Revised Bethesda criteria to undergo genetic testing: direct sequencing of DNA and MLPA were used to examine the entire MLH1, MSH2 and MSH6 coding sequence. Patients were classified as mutation-positive or negative according to the genetic testing result. Results A deleterious MMR mutation was found in 38/302 patients. Median overall survival (OS) was significantly higher in mutation-positive vs mutation-negative patients (102.6 vs 77.7 months, HR:0.63, 95%CI:0.46–0.89, p = 0.0083). Different types of mutation were significantly related with OS: missense or splicing-site mutations were associated with better OS compared with rearrangement, frameshift or non-sense mutations (132.5 vs 82.5 months, HR:0.46, 95%CI:0.16–0.82, p = 0.0153). Conclusions Our study confirms improved OS for LS-patients compared with mutation-negative CRC patients. In addition, not all mutations could be considered equal: the better prognosis in CRC patients with MMR pathogenic missense or splicing site mutation could be due to different functional activity of the encoded MMR protein. This matter should be investigated by use of functional assays in the future.

An intronic mutation in MLH1 associated with familial colon and breast cancer

Single base substitutions can lead to missense mutations, silent mutations or intronic mutations, whose significance is uncertain. Aberrant splicing can occur due to mutations that disrupt or create canonical splice sites or splicing regulatory sequences. The assessment of their pathogenic role may be difficult, and is further complicated by the phenomenon of alternative splicing. We describe an HNPCC patient, with early-onset colorectal cancer and a strong family history of colorectal and breast tumors, who harbours a germ line MLH1 intronic variant (IVS9 c.790 +4A>T). The proband, together with 2 relatives affected by colorectal-cancer and 1 by breast cancer, have been investigated for microsatellite instability, immunohistochemical MMR protein staining, direct sequencing and Multiplex Ligation-dependent Probe Amplification. The effect of the intronic variant was analyzed both by splicing prediction software and by hybrid minigene splicing assay. In this family, we found a novel MLH1 germline intronic variant (IVS9 c.790 +4A>T) in intron 9, consisting of an A to T transversion, in position +4 of the splice donor site of MLH1. The mutation is associated with the lack of expression of the MLH1 protein and MSI in tumour tissues. Furthermore, our results suggest that this substitution leads to a complete skip of both exon 9 and 10 of the mutant allele. Our findings suggest that this intronic variant plays a pathogenic role.

MSH2 missense mutations and HNPCC syndrome: pathogenicity assessment in a human expression system.

Hereditary Non‐Polyposis Colorectal Cancer (HNPCC) is associated with germline mutations in one of several MisMatch Repair (MMR) genes. An increasing proportion (20–25%) of the reported MSH2 variants consists of single amino‐acid substitution with uncertain disease‐causing significance. The present study was undertaken to functionally characterize 3 MSH2 nontruncating variants: p.Gly162Arg (c.484G>C), p.Asp167His (c.499G>C) and p.Arg359Ser (c.1077A>T). Missense alterations, were assessed in a human system for expression/stability and for the ability to heterodimerize with MSH6 and correctly localize into the nucleus. Functional assays results were correlated with clinical and genetic features indicative of HNPCC as MicroSatellite‐Instability (MSI), abnormalities of MMR gene expression in tumour tissue (IHC) and familial history. p.Gly162Arg and p.Arg359Ser variants showed a clearly decreased expression level of the MutSá complex and were associated with an abnormal subcellular localization pattern, which can be suggestive of an incorrect MSH2/MSH6 heterodimerization. Functional analysis results were supported by MSI and IHC data and by familial cancer history. The subcellular localization assay, performed in a human expression system, classifies as pathogenetic two MSH2 nontruncating alterations providing a useful tool in genetic testing programs.

Incidence and risk of cardiotoxicity in cancer patients treated with targeted therapies

BACKGROUND Cardiotoxicityis a serious side effect of molecularly targeted agents. The purpose of this study was to evaluate the incidence and Relative Risk (RR) of developing all-grade and high-grade cardiotoxicity in patients with solid tumors receiving targeted agents through a revised meta-analysis of available clinical trials. METHODS The scientific literature regarding cardiotoxicity was extensively analyzed using MEDLINE, PubMed, Embase and Cochrane Central Register of Controlled Trials (CENTRAL). Eligible studies were selected according to PRISMA statement. Summary incidence, RR, and 95% CIs were calculated using random-effects or fixed-effects models based on the heterogeneity of selected studies. RESULTS Our search yielded a total of 4998 clinical studies; of them, 31 trials were finally considered for this meta-analysis. A total of 28,538 patients were included; 7995 of these patients had breast cancer (28%), 6151 (22%) prostate cancer and 14,392 (50%) were treated for other malignancies. The highest RR of high-grade events was observed with Vandetanib (RR=7.71, 95% CI 1.04-56.99), followed by Ramucirumab (RR=5.0) and Aflibercept (RR=4.1). Grouping by drug category, the highest incidence of high-grade cardiotoxicity was shown by anti-VEGFR-TKIs (RR 5.62, 95% CI 1.49-21.24) and anti-VEGF mAbs/VEGF-trap (RR 1.82, 95% CI 1.24-2.69). Grouping by tumor type, the highest incidence of cardiotoxicity was observed in thyroid cancer (8%), followed by gastric cancer (4%). CONCLUSIONS Treatment with targeted agents in cancer patients is correlated with a significant increase in the risk of cardiotoxicity. Frequent clinical monitoring should be emphasized when using these and newer biological agents.

Role of Vascular Endothelial Growth Factor (VEGF) and VEGF-R Genotyping in Guiding the Metastatic Process in pT4a Resected Gastric Cancer Patients

In radically resected gastric cancer the possibility to predict the site of relapse could be clinically relevant for the selection of post-surgical management. We previously showed that specific tumour integrins genotypes are independently associated with either peritoneal or hematogenous metastases (ITGA and ITGV). Recently VEGF and VEGF-R polymorphisms have been demonstrated to potentially affect tumour angiogenesis and the metastatic process in gastric cancer. We then investigated the role of VEGFs and VEGF-R genotyping in determining either peritoneal carcinosis or hematogenous metastases in radically resected gastric cancer patients. Tumour genotyping for integrins (ITGA and ITGV) was also performed according to our previous findings. Genotyping for VEGF-A, VEGF-C, VEGFR-1,2,3 and ITGA and ITGV was carried out on pT4a radically resected gastric tumours recurring with either peritoneal-only carcinosis or hematogenous metastases. 101 patients fulfilled the inclusion criteria: 57 with peritoneal carcinomatosis only and 44 with hematogenous spread only. At multivariate analysis, intestinal histology and the AC genotype of rs699947 (VEGFA) showed to independently correlate with hematogenous metastases (p = 0.0008 and 0.008 respectively), whereas diffuse histology and the AA genotype of rs2269772 (ITGA) independently correlated with peritoneal-only diffusion (p = <0.0001 and 0.03 respectively). Our results seem to indicate that combining information from genotyping of rs699947 (VEGFA, AC), rs2269772 (ITGA, AA) and tumour histology could allow clinicians to individuate gastric cancer at high risk for recurrence either with peritoneal or hematogenous metastases. The selection tool deriving from this analysis may allow an optimal use of the available treatment strategies in these patients.

Effectiveness of the crcapro program in identifying patients suspected for HNPCC

Subjects affected by hereditary non‐polyposis colorectal cancer exhibit a high susceptibility to colon and extracolonic tumours, due to MMR gene defects. Revised Bethesda criteria are used to select patients as candidates for genetic tests. Recently, the crcapro model has been developed, based on family history of colorectal and endometrial cancers. Our study aims to evaluate the reliability of crcapro in identifying mutation carriers. We used the crcapro program to evaluate carrier probability risk in 99 patients fulfilling Amsterdam or Bethesda guidelines. MLH1 and MSH2 were studied by direct sequencing in all the 99 patients, and the study of microsatellite instability and of MMR proteins expression was performed. Nine MLH1 and nine MSH2 germline mutations were identified. Five out of the nine patients with MLH1 mutation showed a crcapro risk evaluation of less than 20%. The same happened for four out of nine patients with MSH2 mutation. Of the 17 patients with an estimated risk of more than 80%, only four harboured a mutation, all in the MSH2 gene. The highest risk calculated by the crcapro system in the nine carriers of a MLH1 mutation has been 31.7%. In our experience, the crcapro program sensitivity and specificity appears to be low but needs to be further evaluated in larger samples.