Clinical characteristics and outcomes of phase I cancer patients with CCNE1 amplification: MD Anderson experiences

Cyclin E is frequently encoded by CCNE1 gene amplification in various malignancies. We reviewed the medical records of patients with solid tumors displaying CCNE1 amplification to determine the effect of this amplification for future therapeutic development. We reviewed the medical records of patients with advanced solid tumors harboring CCNE1 amplification who were seen at the phase I clinic between September 1, 2012, and December 31, 2019. Among 79 patients with solid tumors harboring CCNE1 amplification, 56 (71%) received phase 1 clinical trial therapy, 39 (49%) had 3 or more concurrent genomic aberrances, and 52 (66%) had a concurrent TP53 mutation. The median overall survival (OS) after patients’ initial phase I visit was 8.9 months and after their initial metastasis diagnosis was 41.4 months. We identified four factors associated with poor risk: age < 45 years, body mass index ≥ 25 kg/m2, presence of the TP53 mutation, and elevated LDH > upper limit of normal. In patients treated with gene aberration-related therapy, anti-angiogenic therapy led to significantly longer OS after their initial phase I trial therapy than those who did not: 26 months versus 7.4 months, respectively (P = 0.04). This study provided preliminary evidence that CCNE1 amplification was associated with frequent TP53 mutation and aggressive clinical outcomes. Survival benefit was observed in patients who received antiangiogenic therapy and gene aberration-related treatment, supporting the future development of a personalized approach to combine gene aberration-related therapy with antiangiogenesis for the treatment of advanced malignancies harboring CCNE1 amplification.

Phase I trial treatment and evaluation. The decision of whether to enroll an eligible study patient in a phase I clinical trial depended on protocol availability and the discretion of the treating physician. Tumor responses were evaluated according to the Response Evaluation Criteria in Solid Tumors (RECIST) (version 1.1) 21 . All patients were followed until death or censored on August 21, 2020. Progression-free survival (PFS) was defined as the time from study entry to the date of first objective documentation of progressive disease, date of death, or censor date. Overall survival (OS) was defined as the time from the date of the first visit to the CCTT (OS-phase I), or as the date of the initial metastasis diagnosis (OS-metastasis) to the date of death or censor date, regardless of whether they received a phase I trial therapy. The phase 1 clinical trial therapy was considered to be a matched therapy if the patient received one or more agents targeting an actionable genetic aberration or protein downstream from it, such as HER2 antibodies for HER2 overexpression or amplification 22,23 or a mitogenactivated protein kinase (MEK) inhibitor for a BRAF mutation 24 . Statistical analyses. Categorical data were summarized with use of frequencies. PFS and OS curves were plotted by using the Kaplan-Meier method and were compared by using log-rank tests. A Cox proportional hazard model was used to determine prognostic. All tests were two-sided and considered significant when P values were less than 0.05. Statistical analyses were performed with use of the SPSS software program (version 25; IBM Corporation).

Discussion
In this retrospective review of cancer patients in a designated phase I clinical trial, several observations have been made. First, CCNE1 amplification was detected in a broad spectrum of human malignancies, more frequently in high-grade malignancies, which was in consistence with its biological functions in cell proliferation and differentiation. CCNE1 amplifications have been observed in more than 7.5% of many tumor types: uterine, breast, www.nature.com/scientificreports/ ovarian, pancreatic, bladder, gastric, esophageal, lung cancers and sarcoma 8 . Cyclin E1 controls the transition of quiescent cells to the cell cycle, as well as epithelial-mesenchymal transitions. Overexpression of cyclin E1 induces replication stress, forcing premature entry into the S phase, and functioning as a key factor for c-myc driven tumorigenesis 25 . Activated cycle E1 further upregulates its own expression by phosphorylating Rb and releasing more E2F independent of mitogenic stimuli. This positive feedback network drives the mitotic transition from the G1 to the S phase, resulting in enhanced cellular proliferation and tumor progression 26 . This aggressive phenotype partly contributes to poor survival in patients with metastatic malignancies harboring CCNE1 amplification, in addition to CCNE1 amplification mediated resistance to chemotherapy 8 . Secondly, we observed a clinical benefit in PFS and OS in patients who were treated with matched therapy targeting concurrent mutations in this cohort of patients. We do not yet have investigational agents for direct inhibition of CCNE1 since the oscillations of the cyclins through the cell cycle are necessary for cell proliferation 27 . However, we investigated whether we would be able to target other concurrent genetic mutations or develop novel strategies to treat CCNE1 amplified cancer cells via synthetic lethal interaction other therapeutics, such as Wee1 kinase or CDK2 kinase inhibition [28][29][30] . Data observed from this retrospective study did support the  www.nature.com/scientificreports/ www.nature.com/scientificreports/ use of targeting concurrent genetic mutations in patients with advanced malignancies harboring undruggable genetic aberrations. Another interesting finding we have observed in this study was that antiangiogenic therapy provided significantly better OS in patients with CCNE1 amplified malignancies. This result might be due to frequent concurrent TP53 mutations in these patients. Among many candidate pathways, the vascular endothelial growth factor (VEGF) pathway serves an important survival function in cancer cells with a mutated TP53 31,32 . TP53 mutations in tumor cells increased the level of hypoxia-induced factor-1α (HIF-1α) and augmented HIF-1α-dependent transcriptional activation of the VEGF gene in response to hypoxia 33 . Cancer cells with mutated p53 have accelerated tumor growth associated with increased VEGF expression and neovascularization 34,35 , which serves an important survival pathway, resulting in a therapeutic advantage of VEGF-inhibition in patients with p53 mutant malignancies, which was supported by our finding that VEGF inhibition-based therapies led to significantly longer PFS in patients with a mutated TP53 than in patients with wild-type TP53 [36][37][38][39] .
As for the use of the therapeutic strategy of synthetic lethality, it was noted that in a phase II clinical of a Wee1 kinase inhibitor adavosertib in patients with CCNE1 amplified advanced solid tumors (NCT03253679), 7 PRs were observed, for an overall response rate of approximately 26%. The median PFS was 4 months and 1-year OS was 55%. In 13 patients with measurable high-grade serous ovarian cancer, 5 (39%) achieved PR and 8 (62%) had SD ≥ 6 months/PR 40 . These results have encouraged us to investigate the development of novel effective therapies by combining targeted therapies against concurrent genetic mutations such as anti-angiogenesis with treatments based on synthetic lethality strategies such as Wee1 kinase inhibition. In addition, the introduction of immunotherapy to these potentially promising therapeutics targeting patients with CCNE1 amplified malignancies warrants further consideration.  www.nature.com/scientificreports/ Finally, we conducted multivariate analyses that revealed four prognostic factors for poor outcomes in this cohort of patients after metastatic disease was initially diagnosed: age < 45 years old, BMI ≥ 25 kg/m 2 , presence of the TP53 mutation, and elevated LDH. Patients in the low-risk group with the risk factor score ≤ 1 had a significantly longer median OS of 55.1 months (95% CI 34.5-75.7 months) after they were initially diagnosed to have metastatic disease than those in the high-risk group with the risk factor score > 1 who had a median OS of 30.2 months (95% CI 15.1-45.3 months). The parameters held true when we applied this model to patients who were initially seen at the phase I trials program but did not receive phase I trial therapy at the CCTT. These prognostic factors have been confirmed to be associated with poor outcome in other reports 38,39,41 .
In considering the clinical importance and relevance of our observations, several limitations should be beard in mind. First, as it is a retrospective study, the selection bias of patient referral to our phase I clinical trials program may limit the generalizability of our findings. Second, small sample sizes in the subgroup analyses may limit the validity of these statistical assessments on other prognostic factors such as the status of concurrent mutations. Finally, the actual molecular profiles were performed at various times and at different CLIA-certified molecular diagnostic laboratories and used next-generation sequencing for different sets of genes, which made it difficult to analyze status of concurrent genetic aberration. Therefore, the findings from this retrospective study should be considered preliminary evidence only for the purpose of hypothesis generation, and require further validation in larger prospective studies.
In summary, the current study, to the best of our knowledge, was the first to analyze clinical outcome in patients with various advanced CCNE1 amplification cancers who were referred to a designated phase I trial clinic. Our study showed that CCNE1 amplification frequently occurred in high-grade adenocarcinoma and that more than 90% of CCNE1 amplification coexisted with other genomic aberrances. Matched therapy targeting concurrent mutations might provide clinical benefit. Multivariate analyses showed four poor prognostic factors: age < 45 years old, BMI ≥ 25 kg/m 2 , presence of TP53 mutation, and abnormally elevated LDH, which was able to classify patients with metastatic malignancies harboring CCNE1 amplification into either the low-risk group with ≤ 1 risk factor or the high-risk group with > 1 risk factor. Our preliminary data showed the role of concurrent genetic mutations, synthetic lethality and antiangiogenesis on patient survival, which may support future development of promising strategic therapeutics for patients with metastatic malignancies harboring CCNE1 amplification, as outline in Fig. 5.

Figure 5.
A sketch of therapeutic development for CCNE1 amplified malignancies. Cycle E overexpression due to CCNE1 amplification and concurrent mutant p53 due to TP53 mutation promote progression from the G1 phase into S phase, providing therapeutic opportunity through synthetic lethality, chemotherapy, targeted therapy and radiation to enhance mitotic catastrophe and apoptosis.