Role of P 53 and Ki-67 as Prognostic Factors in Ovarian Cancer : Systematic Review and Pooled-Analysis

Introduction: Ovarian Cancer (OC) is the most lethal cancer among gynecological malignancies. In the last years, several studies clarified that OC is characterized by different clinical entities that share only the anatomic site. The aim of this review is the evaluation of Ki-67 and p53 as prognostic factors in OC. Methods: By searching Pubmed and abstracts from major cancer meetings, we selected clinical trials within the timeframe 2000-2016. We evaluated all data retrievable and we performed a pooledanalysis by random effect model. The endpoints were the Overall Survival (OS) in terms of Hazard Ratios (HRs) of survival outcomes. Results: We observed a more severe outcome for patients over-expressing Ki-67. However, the overall data of our PFS meta-analysis did not reach the statistical significance (HR 1.98 CI 0.60-6.52, p=0.26). The indirect comparison of p53 studies, here reported, highlighted a possible correlation between patients OS and p53 status. Conclusions: We hypothesize that both Ki-67 and p53 alterations retain a prognostic role in terms of OS in OC. Nicoletta Staropoli1, Domenico Ciliberto1, Eleonora Iuliano1, Antonina Maria De Angelis1, Angela Salvino1, Pierfrancesco Tassone2, Maria Teresa Di Martino2* and Pierosandro Tagliaferri1* 1Department of Medical Oncology, Magna Graecia University, Italy 2Department of Translational Medical Oncology Units, Magna Graecia University, Italy Di Martino MT and Tagliaferri P, et al., Journal of Cancer Clinics Remedy Publications LLC. 2018 | Volume 1 | Issue 1 | Article 1002 2 in ovarian cancer management [10]. However, many studies evaluated the prognostic and predictive role of p53 and BRCA mutations [7,11,12]. Specifically, BRCA mutations represent a biomarker predictive of sensitivity to PARP inhibitors treatment, approved by EMA for the use as single agent in the maintenance setting after a platinumbased treatment in BRCA1/2 mutation carrier patients [8,9,10]. Moreover, based on the systematic review and meta-analysis of the impact of Pegylated Liposomal Doxorubicin (PLD) compared to no-PLD-based regimens in ovarian cancer treatment, we observed no significant advantages in terms of OS, RR or Ca125-response [13]. Patients and Methods Study design and data extraction Systematic review of the Scientific Literature was performed by interrogating major dedicated search engines (PubMed). In order to select and collect homogeneous studies, two investigators (N.S. and E.I.) examined each trial, independently [14]. All discrepancies were resolved by an arbiter (P.T.). From selected trials, the following variables were analyzed and efficacy results were extracted: First author, number of patients enrolled, year of publication, treatment schedule, involved pathway, and so on. Efficacy endpoints here specified, including OS and Progression Free Survival (PFS) were analyzed. Data extraction was performed in accord to the PRISMA statement. The role of several genes (such as p53, Ki-67) was explored. The selected keywords were: “ovarian”, “ovary”, “tumor”, “cancer”, “advanced”, “metastatic”, “therapy”, “Ki-67”, “p53” in different combinations. The ‘related articles’ function and references retrieved from articles were used to perform the search of all related studies, abstracts and citations. From selected trials identified, the following variables were evaluated and efficacy results were extracted: first author, number of patients enrolled, year of publication, treatment schedule, efficacy endpoints (OS, PFS, RR) if analyzed. Data extraction was in accord to the PRISMA guidelines. Results TP53 Study selection and characteristics: PRISMA chart related to published data is described in (Figure 1) considering a timeframe (2000-2016). The used key words were “epithelial”“ovarian cancer”, “p53” and “prognostic and/or predictive factor”. Starting by 2386 articles identified using the pre-specified key words only 6 studies were included in final analyses [15,16]. All published phases II and III studies are reported in (Table 1). Moreover, 5 studies evaluated role of several p53 SNPs in term of OS. As previously reported and described, by the indirect comparison of p53 studies, it was possible to hypothesize a possible favourable correlation between patients OS and p53 status. Ki-67 Study selection and characteristics: The PRISMA chart related to published studies is described in (Figure 2). We considered a timeframe (2000-2016). The used key words were epithelial ovarian cancer, Ki-67 and prognostic and/or predictive factor. Starting by 147 articles identified using the pre specified key words only 5 studies were included in final analyses [17,18]. We excluded reviews, comment and meta-analyses. All published phases II and III studies are reported in (Table 2). OS and PFS analyses For a total of 5 studies, 617 patients were enrolled in this metaanalysis. All patients presented an advantage stage (III-IV FIGO Figure 1: PRISMA chart related to published data in the timeframe 2000-2016. Only 6 studies were included in final analyses. Figure 2: PRISMA chart related to published data in the timeframe (2000-2016). Only 5 studies were included in final analyses. Di Martino MT and Tagliaferri P, et al., Journal of Cancer Clinics Remedy Publications LLC. 2018 | Volume 1 | Issue 1 | Article 1002 3 stage). All patients received almost one line of platinum-based chemotherapy. In terms of OS, we showed that the over expression of Ki-67 (> 30%) indicated a poor prognosis (HR 3,91 Cl 1,07-14.19; p=0.03). Also concerning PFS analyses, we reported a more severe outcome for patients over-expressing Ki-67. In particular, the study of Battista et al. also showed in terms of PFS a HR of 11.5 [19]. However, the overall data of our PFS meta-analysis did not reach the statistical significance (HR 1.98 CI 0.60-6.52, p=0.26) (Figures 3,4).