Comprehensive Characterization of the Vascular Effects of Cisplatin-Based Chemotherapy in Patients With Testicular Cancer

Background Cisplatin-based chemotherapy increases the risk of cardiovascular and renal disease. Objectives We aimed to define the time course, pathophysiology, and approaches to prevent cardiovascular disease associated with cisplatin-based chemotherapy. Methods Two cohorts of patients with a history of testicular cancer (n = 53) were recruited. Cohort 1 consisted of 27 men undergoing treatment with: 1) surveillance; 2) 1 to 2 cycles of bleomycin, etoposide, and cisplatin (BEP) chemotherapy (low-intensity cisplatin); or 3) 3 to 4 cycles of BEP (high-intensity cisplatin). Endothelial function (percentage flow-mediated dilatation) and cardiovascular biomarkers were assessed at 6 visits over 9 months. Cohort 2 consisted of 26 men previously treated 1 to 7 years ago with surveillance or 3 to 4 cycles BEP. Vasomotor and fibrinolytic responses to bradykinin, acetylcholine, and sodium nitroprusside were evaluated using forearm venous occlusion plethysmography. Results In cohort 1, the percentage flow-mediated dilatation decreased 24 h after the first cisplatin dose in patients managed with 3 to 4 cycles BEP (10.9 ± 0.9 vs. 16.7 ± 1.6; p < 0.01) but was unchanged from baseline thereafter. Six weeks after starting 3 to 4 cycles BEP, there were increased serum cholesterol levels (7.2 ± 0.5 mmol/l vs. 5.5 ± 0.2 mmol/l; p = 0.01), hemoglobin A1c (41.8 ± 2.0 mmol/l vs. 35.5 ± 1.2 mmol/l; p < 0.001), von Willebrand factor antigen (62.4 ± 5.4 mmol/l vs. 45.2 ± 2.8 mmol/l; p = 0.048) and cystatin C (0.91 ± 0.07 mmol/l vs. 0.65 ± 0.09 mmol/l; p < 0.01). In cohort 2, intra-arterial bradykinin, acetylcholine, and sodium nitroprusside caused dose-dependent vasodilation (p < 0.0001). Vasomotor responses, endogenous fibrinolytic factor release, and cardiovascular biomarkers were not different in patients managed with 3 to 4 cycles of BEP versus surveillance. Conclusions Cisplatin-based chemotherapy induces acute and transient endothelial dysfunction, dyslipidemia, hyperglycemia, and nephrotoxicity in the early phases of treatment. Cardiovascular and renal protective strategies should target the early perichemotherapy period. (Clinical Characterisation of the Vascular Effects of Cis-platinum Based Chemotherapy in Patients With Testicular Cancer [VECTOR], NCT03557177; Intermediate and Long Term Vascular Effects of Cisplatin in Patients With Testicular Cancer [INTELLECT], NCT03557164)

T esticular cancer is the most common malignancy in men aged 20 to 40 years, and cisplatin-based chemotherapy with bleomycin, etoposide, and cisplatin (BEP) achieves a cure for almost all patients (1,2). However, this treatment is associated with cardiovascular disease (CVD), including myocardial infarction, thrombosis, and nephrotoxicity (3,4).
Adverse effects on the endothelium appear to be a central pathophysiological mechanism (5), and perturbations in metabolic and inflammatory parameters may also be important (6,7). A retrospective epidemiological study of >15,000 patients demonstrated a 5-fold increase in standardized cardiovascular mortality following cisplatinbased chemotherapy that was confined primarily to the first year following treatment (1). This challenges the preconception that cardiovascular risk in testicular cancer survivors is a late phenomenon (3,8).
Historically, the prospective evaluation of vascular effects of cisplatin-based chemotherapy has been hindered by assessments in heterogeneous groups with differing cardiovascular risk, cancer types, and treatment regimens. Furthermore, examination of immediate effects has been limited and usually without longitudinal assessment (9)(10)(11). Understand-      Patients with a recent diagnosis of testicular/retroperitoneal germ cell cancer and orchidectomy #8 weeks prior scheduled for cisplatin-based chemotherapy or active surveillance were recruited. Participants were stratified into 3 groups by management: 1) active surveillance for stage  Blood samples were simultaneously drawn from each   were analyzed as described previously (18). Net t-PA and PAI-1 release were defined as the product of the infused forearm plasma flow and the concentration difference between infused and noninfused arms (18). Previous studies demonstrated that 8 subjects per group provides sufficient power to detect an approximately 20% change in FBF at 5% significance (19,20).   Table 2). This remained numerically greater than baseline thereafter, but was not statistically significant. There were trends toward  Abbreviations as in Table 2.   Similar results were found for ERK 1/2 phosphorylation (n ¼ 5; p ¼ 0.026). t-PA mRNA expression decreased in cells exposed to cisplatin (n ¼ 7;

Continued on the next page
p ¼ 0.014), whereas PAI-1 mRNA expression was unchanged (n ¼ 7; p ¼ 0.122) ( Figure 5). Endothelial dysfunction is a key feature in de novo thrombosis and rupture of pre-existing atherosclerotic plaque (25). As such, it is an important component to the risk of thrombotic cardiovascular events in the early period following cisplatin-based chemotherapy.
In our cell-based study, exposure of HAECs to cisplatin was associated with decreased t-PA mRNA expression and activation of Akt and ERK 1/2. In previous work, cisplatin has been shown to reduce endothelial cell survival and induce apoptosis in human dermal microvascular endothelial cells (HMEC-1) (26) and in the vasa nervorum in rats (27). These data support the hypothesis that cisplatin induces direct endothelial toxic effects resulting in increased stress kinase signaling and a propensity for thrombosis via reduction in the capacity for endogenous fibrinolysis.
Although adverse thrombotic effects are most frequently observed in the early period following chemotherapy, cisplatin is detectable in serum for several years after treatment (28). This chronic exposure may provoke low-grade endothelial stimulation and a consequent pro-atherogenic environment (28,29). Our subsequent assessment was with forearm venous occlusion plethysmography, the gold standard for assessing endothelial function (12 (32).
Most studies evaluating the risk of CVD after cisplatin-based chemotherapy have focused on the later period after treatment (3,8,33,34). In the 10 to 19 years following treatment, a 5-to 7-fold increased risk of CVD has been reported compared with patients managed with surveillance or the general population (8,33). Larger studies of up to 2,700 patients demonstrate a more modest 1.5-to 2-fold increased risk of CVD at 10 to 18 years after treatment (3,34  appears to be the time period of increased cardiovascular risk. While these effects predominantly occur in the early period after treatment, cisplatin is detectable in serum several years after treatment (28). It remains possible that chronic exposure to low levels of cisplatin may cause low-grade endothelial stimulation that contributes to the pathophysiology underlying cardiovascular events occurring more than a decade after initial exposure (8,28).
Nephrotoxicity is a major, dose-limiting side effect of cisplatin that affects 20% to 40% of patients (4,36).
We found evidence of early nephrotoxicity following 3 to 4 cycles of BEP as illustrated by increased serum cystatin C (37). Importantly, this returned to baseline by 3 months and, although there was an absolute decrease in eGFR, the change in this less-sensitive measure did not reach statistical significance.  Patients with testicular cancer managed with cisplatin-based chemotherapy or surveillance were enrolled. Endothelial function was assessed by % flow-mediated dilatation over 9 months in patients with a recent diagnosis, or venous occlusion plethysmography in patients managed 1 to 7 years previously. Acute and transient endothelial dysfunction was observed 24 h after cisplatin-based chemotherapy, and endothelial function was not impaired 1 to 7 years after treatment. In the left panel, the blue and red shaded rectangles refer to duration of chemotherapy treatment (solid blue ¼ 1 cycle, light blue ¼ 2 cycles; solid red ¼ 3 cycles, light red