Assessment of pazopanib-related hypertension, cardiac dysfunction and identification of clinical risk factors for their development

Background Antineoplastic therapy with the tyrosine kinase inhibitor pazopanib in patients with advanced/metastatic renal cell carcinoma (mRCC) has been associated with hypertension (HTN), cardiomyopathy, and cardiac dysrhythmias. We therefore assessed the cardiovascular (CV) risk with pazopanib in a clinical setting. Methods Medical records of 35 antineoplastic-naïve mRCC patients newly started on pazopanib were retrospectively reviewed at a single academic medical center. Assessment of the hypertensive response and adverse cardiac events associated with pazopanib was the primary objective. Outcomes were defined using the National Cancer Institute’s Common Terminology Criteria for Adverse Events v4.0. Potential clinical risk factors were investigated with univariate and multivariable logistic regression. Results Pazopanib-induced HTN was observed in 57% of patients. Median maximal systolic blood pressure (SBP) during pazopanib treatment was 167.5 mmHg with median time to event of 24.5 days. New-onset HTN occurred in 6/14 (43%) patients. Baseline SBP > 130 mmHg (odds ratio [OR]: 5.32; 95% confidence interval [CI]: 0.94-29.99; p = 0.058) and ACEi/ARB use (OR: 4.88; 95% CI: 1.05 22.84; p = 0.044) were risk factors for pazopanib-induced HTN. When HTN was excluded, 34% of patients developed a CV adverse event. Age ≥ 60 years (OR: 8.72; 95% CI: 0.74-513.26; p = 0.105) trended towards being a predictor for a non-HTN CV adverse event. Conclusions Our findings suggest that pazopanib has a broad CV toxicity profile in treatment-naïve mRCC patients headlined by a rapid and striking hypertensive response. More intensive BP control prior to starting pazopanib and standardization of CV surveillance particularly in older patients may optimize oncologic care while minimizing CV risk.


Background
Vascular endothelial growth factor (VEGF) signaling pathway inhibitors (VSPI) have known efficacy in multiple malignancies by inhibiting tumor angiogenesis, but are increasingly being recognized as cardiotoxins. Smallmolecule targeted VEGF receptor (VEGFR) tyrosine kinase inhibitors (TKI) have significantly improved outcomes in advanced/metastatic renal cell carcinoma (mRCC) as evidenced by 6 and 14 month increases in median progression free survival and overall survival, respectively, with sunitinib compared to earlier first line agents for mRCC [1,2]. Pazopanib is a newer oral angiogenesis inhibitor targeting VEGFR-1, −2, and −3, platelet-derived growth factor receptor (PDGFR)-α and -β, and c-KIT and has been a first-line agent for mRCC since 2009 [3][4][5]. Potential expansion of pazopanib's use in various pediatric and adult malignancies is currently being investigated [6]. Pazopanib has similar efficacy compared to sunitinib, a similar multi-targeted VSPI that preceded it and whose cardiotoxic effects are described most often within this class [7,8]. Within its class, pazopanib's favorable overall side effect profile and cost-effectiveness have made it an appealing option for physicians and patients [9][10][11][12][13]. Consideration of these factors suggests that an increasingly higher number of patients with mRCC will be treated with pazopanib.
Roughly 63,000 patients in the United States are diagnosed with renal cancer annually [25]. Given that the median age of diagnosis of RCC is 64 years, many of them have an increased risk or may already have preexisting CV disease prior to initiating targeted VSPI treatment such as pazopanib [26]. Development of clinically significant HTN can result in morbidity and pazopanib dose reduction or cessation, thus limiting the overall efficacy of cancer treatment. Our objective was to characterize the extent of CV toxicity associated with pazopanib and the risk factors for its development in an antineoplastic-treatment naïve, real-world mRCC patient population to capture pazopanib's unique CV effects.

Study participants
Cases were selected from 462 consecutive male and female patients, age 18 years or greater, with a diagnosis of mRCC. International Classification of Diseases -9 and 10 (ICD-9/10) diagnosis codes were used to identify cases. All patients had been treated with pazopanib within the Ohio State University Wexner Medical Center (OSUWMC) health system at some point during the period 12/01/2009 to 08/01/2016 and had at least two follow-up visits with an OSUWMC clinician during pazopanib therapy. Cases were excluded if baseline blood pressure (BP) was missing, pazopanib therapy was stopped fewer than 7 days after initiation, or if the patient underwent treatment with any other systemic antineoplastic agent prior to pazopanib exposure. This excluded 427 patients and the 35 remaining patients comprised the final cohort for this study. All 35 patients were followed-up until either death occurred or until their last encounter with an OSUWMC clinician. Follow up was completed in August 2016. The study was approved by the Ohio State University (OSU) Cancer Institutional Review Board.
Baseline patient characteristics were captured using OSUWMC electronic medical records (EMR). This included past medical history elements and medication lists provided at each oncologic-related visit. Study entry date was set as the time of first pazopanib order placed in the EMR. Baseline characteristics included age at pazopanib start date, Eastern Cooperative Oncology Group (ECOG) performance status, tumor histology, starting pazopanib dose, preexisting comorbidities, medications, smoking status, and body mass index (BMI). Cardiovascular comorbidities of interest were HTN, diabetes, dyslipidemia, renal insufficiency defined as a glomerular filtration rate less than 60 ml/min/1.73 m 2 , coronary artery disease, peripheral arterial disease, congestive heart failure, left ventricular dysfunction, cardiac dysrhythmias, cerebrovascular disease, and thromboembolic disease. We identified use of angiotensin-converting enzyme inhibitors (ACEi), angiotensin-receptor blockers (ARBs), beta-blockers (BBs), calcium-channel blockers (CCBs), diuretics, statins, and metformin at or before study entry date. Medications were followed longitudinally for the entirety of the study period using medication lists available with every oncology office visit within the OSUWMC James Cancer Hospital. Criteria used to assess ongoing pazopanib treatment were a minimum of two related visits in which pazopanib appeared in the medication list. Validation by manual chart review of patient medication lists and oncologic provider documentation in the EMR was performed for each patient to ensure accuracy of pazopanib treatment dates.

Cardiovascular data review
The primary source for clinical variables was EMR data entered by trained healthcare professionals for clinical purposes within the OSUWMC system. Baseline systolic (S) and diastolic (D) BP was determined using the mean value of measurements obtained at each oncologic office visit in the preceding 90 days of pazopanib start date for each patient. SBP and DBP after pazopanib initiation was determined using the mean value of measurements obtained at each subsequent oncology-related office visit which at minimum included two visits (one at two weeks post-pazopanib initiation and one at four weeks post-pazopanib initiation). Baseline left ventricular ejection fraction (LVEF) was available for 25 out of the total 35 patients in this study. Echocardiography data was obtained from final reports that were only available after an official interpretation was entered by an expert cardiologist. Baseline LVEF determination was with conventional 2-dimensional (2D) echocardiography using the Simpson biplane technique, according to the American Society of Echocardiography guidelines [27]. Baseline electrocardiographs (ECG) were available for review within the EMR in 29 patients. An expert cardiologist had previously confirmed each ECG interpretation. Data extracted were the corrected QT-interval (QTc) derived from Bazett's formula (QTc = QT/√RR) and the QRS duration. ECGs were manually reviewed and excluded if they had features such as an electronically paced rhythm and/or significant intra-ventricular conduction delay that do not allow for accurate QTc measurement.

Definition of outcomes
The development of pazopanib-induced HTN was the primary outcome studied. Patients with or without preexisting HTN could meet criteria for the primary outcome. Patients were considered having preexisting HTN if they met any one of the following criteria prior to the date of pazopanib initiation: (a) HTN documented as a diagnosis in the EMR (b) at least one prescribed medication within the antihypertensive class (c) systolic blood pressure (SBP) greater than or equal to 140 mmHg or a diastolic blood pressure (DBP) greater than or equal to 90 mmHg at least two separate clinical encounters. These parameters were chosen in accordance with the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE v4.0) and Joint National Committee on Prevention, Detection, and Treatment of High Blood Pressure (JNC) 8 [28,29].
For those without preexisting HTN, pazopanibinduced HTN was defined as the occurrence of any one of the above criteria during pazopanib treatment. For patients with preexisting HTN, pazopanibinduced HTN was defined as any one of the following interventions during pazopanib therapy: (a) addition of a new antihypertensive medication, (b) dose escalation of a baseline antihypertensive medication. Severity of HTN was also assigned based on CTCAE v4.0 definitions, graded 1 to 5 according to severity [29]. CV adverse events (AE) were defined in accordance with the CTCAE v4.0 definitions due its universal acceptance in defining AEs in oncologic clinical trials. CV AEs chosen for inclusion in this study were "Hypertension," "Heart Failure," "Electrocardiogram QT corrected interval prolonged," "Atrial flutter," and "Peripheral Ischemia". Table 1 lists all definitions of Grade 2: Stage 1 hypertension (systolic BP 140-159 mmHg or diastolic BP 90-99 mmHg); medical intervention indicated; recurrent or persistent (≥24 h); symptomatic increase by >20 mmHg (diastolic) or to >140/90 mmHg if previously within normal limits; monotherapy indicated Grade 3: Stage 2 hypertension (systolic BP ≥160 mmHg or diastolic BP ≥100 mmHg); medical intervention indicated; more than 1 drug or more intensive therapy than previously used indicated AEs and severity grades (1 to 5, in ascending severity). Any AE with a grade of 3, 4, or 5 was considered a high-grade event in accordance with the CTCAE.

Statistical analysis
Continuous variables are presented as means with standard or as medians with interquartile ranges ([IQR]: 25th-75th percentile.) Categorical variables were compared using Student's t test, Mann-Whitney U test, chi-square test, or Fisher's exact test as appropriate. Univariate logistic regression analysis was performed to estimate odds ratios (ORs) of potential risk factors for the development of pazopanib-induced HTN. Any clinical variables identified in this analysis with P < 0.1 were entered into a multivariable logistic regression model to identify independent factors associated with development of pazopanib-induced HTN. In addition to analyzing SBP, age, and BMI as continuous variables, binary variables were established by dichotomizing SBP (above or below 130 mmHg), age (above or below 60 years), and BMI (above or below 30 kg/m 2 ) for logistic regression analysis. ORs with 95% confidence intervals (CIs) were generated. Overall survival outcomes were assessed using the log-rank test and Kaplan Meier survival estimates. For all comparisons, P < 0.05 was considered significant. All statistical analyses were performed using STATA 14.

Patient characteristics
Baseline demographic and clinical data of the entire cohort and their comparison between patients who did and did not develop pazopanib-induced HTN is illustrated in Table 3. The majority of patients had clear cell tumor histology (80%) and were started on the standard trial dose of pazopanib 800 mg daily (91%). Among the total cohort, more than half of patients had a prior nephrectomy (63%), current or past smoking history (60%), hypertension (60%), ECOG performance status score of 1 (54%), and renal insufficiency (57%). Gender, mean age at pazopanib initiation, and BMI levels were largely similar. A baseline ECOG performance status score of 0 was predominantly observed in the pazopanib-induced HTN group. Compared to patients who did not develop pazopanib-induced HTN, patients in the pazopanib-induced HTN group had a significantly higher mean baseline SBP (130.5 ± 10.9 mmHg vs. 121.7 ± 8.2; P = 0.01) and DBP (74.4 ± 8.5 mmHg vs. 69.7 ± 4.4; P = 0.045) along with a higher proportion of baseline ACEi or ARB usage (55% vs. 20%; P = 0.037) and higher incidence of pazopanib dose reduction during therapy (30% vs. 0%, P = 0.027). Other baseline CV disease and CV risk factors were not significantly different between the two groups. Median length of follow-up was 10 months (IQR: 3.1-19.4 months) for the total cohort and mortality occurred in 15/35 (43%) patients during the study period. Both characteristics were not significantly different between the two groups Table 2.
Description of the Pazopanib-induced hypertensive response and associated risk factors The majority of patients in our cohort (57%) developed pazopanib-induced HTN ( Table 3). The overall median time from pazopanib start date to development of pazopanib-induced HTN was 24.5 days (IQR: 14.5-53.5 days). Of the 14 patients without preexisting HTN, 6 (43%) developed new-onset HTN with the median time to incident HTN of 19 days (range 7-53 days). Preexisting HTN was present in 21 patients, and 14 patients met criteria for pazopanib-induced HTN with a median time to event of 29.5 days (IQR 18-92 days). As Fig. 1 illustrates, there was a significant increase in SBP from baseline to the maximal measured during pazopanib treatment with an overall median SBP after pazopanib exposure 8.2 mmHg higher than baseline in pazopanib-induced HTN patients. A systolic blood pressure increase greater than 10 mmHg from baseline on at least two separate BP measurements during pazopanib therapy was seen in 25/35 (71%) patients in the entire cohort. Among patients in our cohort meeting criteria for pazopanib-induced hypertension, 15/20 (75%) patients had a systolic blood pressure increase greater than 10 mmHg from baseline on at least two separate BP measurements during pazopanib therapy. A total of 26 distinct episodes of either initiation or dose escalation of an antihypertensive occurred. ACEi/ ARBs (46%) and CCBs (27%) accounted for the majority of these (Table 3).

Pazopanib-related cardiovascular Adverse events and associated risk factors
Nearly 70% of patients in our study developed CV toxicity (Fig. 2, Table 5). When HTN was excluded, 12 of 35 patients (34%) still met criteria for developing a CVAE during the course of pazopanib treatment. QTc-interval prolongation was most common among these and represented 23% of all CVAEs. As shown in Fig. 3, pazopanib treatment was strongly associated with prolongation of the QTc-interval with a median increase of 16 ms (p = 0.057) detected in the 24 patients with baseline and treatment ECGs. Among the 7 total patients who had LVEF assessments before and after pazopanib exposure, an absolute decline in LVEF was observed in 5 patients (Figure 4). Among these 5 patients, two developed clinically significant declines in LVEF, defined as greater than 10%. This was not associated with concomitant uncontrolled hypertension at the time of diagnosis of LVEF decline with measured BPs of 114/79  (Table 6). With the exception of age and prior CVA/TIA, no significant differences were found between those who developed a CVAE and those who did not after excluding HTN (Table 7). Age ≥ 60 years was associated with non-HTN CVAE (OR: 8.72; 95% CI: 0.74-513.26; p = 0.105) though did not meet statistical significance as an independent predictor and on exploratory analysis, prior CVA/TIA was an additional risk factor (OR: 8.61; 95% CI: 0.86-infinite; p = 0.067) ( Table 8). Statistical significance was also not maintained on multivariable adjusted logistic regression. There was no significant association between statin (p = 0.568) or beta-blocker (p = 0.714) therapy and survival.

Discussion
The major findings in this study of antineoplastic-naïve mRCC patients newly started on pazopanib includes: (1) a strikingly high proportion (69%) of patients developing a form of CV toxicity ranging from asymptomatic cardiac repolarization abnormalities on ECG to fatal cardiogenic shock; (2) a marked and rapid hypertensive response corresponding to a higher observed rate of high-grade HTN in our cohort than previously reported with pazopanib; and (3) an absolute decline in LVEF  Fig. 1 Median systolic blood pressure before and after pazopanib initiation in patients meeting criteria for pazopanib-induced hypertension (N = 20). Baseline median SBP is at the far left and is equal to 128.6 mmHg. Median maximal SBP is within the middle box and is equal to 167.5 mmHg. Median time to reach maximal SBP was 24.5 days as described in Table 3. Overall median SBP during pazopanib treatment is within the box on the far right and is equal to 136.8 mmHg. Solid line within each box represents the median. Boxes represent the interquartile range. Bars represent the range. SBP: systolic blood pressure after pazopanib exposure in 5/7 (71%) patients who had LVEF assessments before and after treatment.

Pazopanib-induced hypertension
Any grade of pazopanib-induced HTN was seen in 20/35 (57%) patients, exceeding the reported incidence rates of 36-46% [4,19,20]. A marked hypertensive response (>20 mmHg increase in SBP or DBP) was observed in these 20 patients. Notably 85% of the patients in our study met CTCAE v4.0 criteria for grade 3 HTN, which is a significantly higher proportion than the 4-7% incidence reported in earlier phase II/III clinical trials [4,20]. Two contributing factors may explain this discrepancy. First, early pazopanib trials from which much of the data comes from   [30]. After peak BP levels were achieved, we saw a subsequent decline towards baseline. This likely represents the effect of more intensive antihypertensive therapy after treating clinicians recognized pazopanib-induced HTN. A similar pattern has been observed in prior studies involving multiple agents within the VSP inhibitor class [30][31][32]. Not surprisingly, preexisting hypertension has been found to be a risk factor for VSPI-induced HTN [19,31,33]. Our data are consistent with these previous findings, with an association existing between the development of pazopanib-induced HTN and presence of a baseline prehypertension. Of note, 17/20 (85%) patients with preexisting HTN had adequate baseline BP control (<140/90 mmHg) using JNC-8 guidelines and BP targets before initiation of VSPIs proposed in prior studies [28,31]. Despite the vast majority of preexisting HTN patients in our study achieving these targets, they still developed pazopanib-induced HTN with strikingly high magnitudes of BP elevation. Given that mRCC patients have relatively limited life expectancies, acute complications from uncontrolled HTN have historically been of particular concern with VSPI initiation [34,35]. However, advances in the treatment of mRCC have improved survivorship to the point where median overall survival in pazopanib-treated patients is now 22.9 months and improves to 42.5 months in patients with favorable oncologic features [4,36]. In our study, all 6 of the patients requiring pazopanib dose reduction also developed pazopanib-induced HTN. A lower BP target prior to initiating pazopanib Two recent studies have shown improved overall survival rates in mRCC patients on VSPI therapy undergoing concomitant treatment with an ACEi or ARB [37,38]. Our data did not demonstrate a similar ACEi/ARB survival benefit. This may be due to our cohort being limited to only pazopanib-treated mRCC patients given that exploratory analysis from a large secondary pooled analysis of two RCTs also did not find a survival advantage in pazopanib-treated patients on ACEi/ARB therapy [39]. It has been proposed that ACEi/ARBs act synergistically with VSPIs to enhance their antineoplastic effect [38]. It is unclear from a mechanistic standpoint why the same degree of potentiation with pazopanib is not seen and also highlights the need to study VSPI agents individually to better characterize their clinical effects.

Pazopanib-related cardiac toxicity
After excluding HTN, 12/35 patients (31%) still met criteria for developing a CVAE which is significantly higher than what was described in clinical trials but is consistent with a similar CV-focused study in a clinical setting where 13/43 (30%) pazopanib-treated patients developed a non-HTN CVAE [40]. A few differences in each study, however, are worth noting. By design, our cohort of mRCC patients was treatment naïve with no H/O history of, SP status post, AV atrioventricular, PPM permanent pacemaker, SB shortness of breath, LE lower extremity, BNP brain natriuretic peptide, LVEF left ventricular ejection fraction, TTE transthoracic echocardiogram, PEA pulseless electrical activity, HFrEF heart failure with reduced ejection fraction, JVP jugular venous pressure, CMR cardiac magnetic resonance, IV intravenous, GDMT guideline directed medical therapy, HTN hypertension, DM diabetes mellitus, HLD hyperlipidemia, TEE transesophageal echocardiogram, EPS electrophysiology study, DCCV direct-current cardioversion, CAD coronary artery disease, CVA cerebrovascular disease, PAD peripheral arterial disease prior exposure to potentially cardiotoxic antineoplastic agents. This was by design given the uncertainty of potential long-term cardiotoxic effects with some of the novel agents used in the treatment of mRCC. Secondly, because the other investigators had an established CV monitoring protocol for TKI-treated patients, they were able to utilize cardiac biomarkers as a measure of cardiotoxicity and had a higher proportion of patients with LVEF assessments at baseline and during treatment that led to a higher detection rate of low grade HF as defined in Table 1. Conversely, our study included cardiac conduction abnormalities and peripheral ischemia as CVAEs while theirs did not. The overall severity of pazopanib-related CVAEs that we observed appears to be higher by comparison. In particular, we observed two HF events and both were high-grade, with one resulting in death and the other in pazopanib discontinuation (Tables 5, 6) compared to all HF events being grade 1 or 2 severity in this earlier study. The development of high-grade HF in 2/35 (6%) patients in our study is almost six times the rate reported in early clinical trials [20]. Though this may be reflective of the difference in number of study participants, it is also possible the higher rate we observed is related to our cohort consisting of patients outside the clinical trial setting, with a higher burden of comorbidities. Our data suggests that age greater than 60 years may increase the risk for a pazopanib-related non-HTN CVAE. This is not surprising given CV risk increases with age even in healthy adults given the higher prevalence of comorbid conditions such as HTN, DM, and atherosclerotic disease. There may be a potential mechanistic link to this observation given the experimental finding that older mice treated with the TKI imatinib experience more severe cardiotoxicity as a result of age-dependent increase in oxidative stress [41]. It is also worth noting that the most wellrecognized cardio-oncology clinical guidelines specifically recommend increased attention to cardiac function surveillance for patients ≥60 years old treated with anthracyclines and/or trastuzumab given limited data on this population [42]. Given the uncertainty of the full scale of cardiotoxicity with pazopanib and other novel VSPIs, prospective clinical studies assessing the benefit of standardized CV functional assessment in this patient population is warranted. QTc interval prolongation with small molecule TKIs such as pazopanib has been postulated to be related to "off target" blockade of the HERG K+ channel and has the potential to increase the risk of potentially lifethreatening unstable ventricular dysrhythmias [43,44]. We observed a higher proportion of patients developing QTc intervals >500 ms than what was reported in clinical trials (6% vs. <2%). Given that cancer patients are prone to diarrhea-and vomiting-related electrolyte derangements, these may have been identified and corrected at higher rates in the clinical trial setting where study participants are monitored with more frequent lab testing at regular intervals. Another contributing factor may have been a higher degree of concomitant use of QTc-prolonging medications such as antiemetics and psychotropics in our cohort. Of the 9 patients in our study that developed CV toxicity in the form of QTc prolongation, none developed Torsades de Pointes or another life-threatening ventricular dysrhythmia. Though these events are rare (<1% incidence in clinical trials), the significant morbidity and mortality rate they pose warrants regular ECG monitoring during pazopanib treatment.

Limitations
This was a retrospective, observational single center study with information obtained from the EMR. As with all analyses using EMRs, potential introduction of unidentifiable sources of bias warrants consideration. Preexisting HTN may have been affected by gender, age, patient comorbidities, and concurrent medication use. We attempted to address this problem by using multivariate risk adjustment, but unmeasured variables inherently cannot be accounted for in this study design. Variability in hospital coding practices and physician documentation may have resulted in underestimation of some comorbidities. The use of a standardized EMR data extraction template and physician review of medical records was employed to minimize this factor.
The lack of standardization of echocardiographic monitoring of LV function was evident in our study cohort, with a large proportion of patients not having regular surveillance of LV function during pazopanib treatment and 25/35 (71%) patients having a baseline LVEF assessment. This makes it difficult to draw significant conclusions about the echocardiographic data collected though a trend toward LVEF reduction with pazopanib treatment was observed. Variability in physician echocardiograph interpretation was also a source of potential bias. We attempted to minimize this by collecting strictly quantitative data from echocardiography reports. The lack of echocardiographic screening prior to pazopanib CVA/TIA was a perfect predictor for a non-hypertension cardiovascular event. P values based on exact logistic regression initiation in nearly 30% of our total cohort may be reflective of under-recognition of the CV risk pazopanib poses. Considering the marked hypertensive response with pazopanib therapy, implementation of a standardized CV risk assessment protocol that incudes echocardiographic screening for these patients is warranted. The lack of standardized measurement of cardiac biomarkers in our patient cohort is also a limitation as it may have caused under-detection of pazopanib-induced subclinical CV toxicity. Incorporating cardiac biomarker measurement into CV risk assessment and surveillance protocols in clinical practice before and during pazopanib therapy should thus be considered. The size of the patient cohort that was utilized may limit our generalizability. A larger total cohort may have allowed for identification of more predictors of CVAEs and perform more robust survival analysis. By study design, we excluded patients treated with any other systemic antineoplastic agent including TKIs such as sunitinib, sorafenib, cabozantinib or axitinib. Since our focus was to assess CV risk factors and cardiotoxicity associated strictly with pazopanib in the hopes of possibly elucidating mechanistic links specific to pazopanib, this would have introduced a major confounder into our study. However, given the known overlap in receptor affinities among the VSPIs, our findings could potentially be applied to other agents in this class. Lastly, the study population was composed of patients from a single healthcare system. As a result, the level of generalizability is not entirely clear. However, the OSUWMC is a large tertiary care referral center and the population of patients encountered likely resembles most large medical centers.

Conclusions
This is the first study that exclusively examined pazopanib-induced CV effects in antineoplastic-naïve mRCC patients in a clinical setting. Our findings suggest that pazopanib possesses a multifaceted cardiovascular toxicity profile which includes cardiomyopathy ranging from asymptomatic reduction in LVEF to fatal cardiogenic shock, cardiac repolarization disturbances manifested by QTc-interval prolongation, and a striking hypertensive response predominantly within 30 days of starting pazopanib that was associated with pazopanib dose reduction.
Preexisting CV disease has been identified in as many as 35% of renal cell cancer patients in the US [45]. Combined with the fact the cancer survivorship continues to improve with the rapid evolution of targeted therapies such as pazopanib, the intersection between cardiovascular and oncologic disease will likely continue to expand. Standardization of CV risk stratification prior and cardiac surveillance in patients undergoing treatment with pazopanib and other VSPIs can optimize oncologic care while minimizing potentially avoidable CV risk. The findings presented here are hypothesis generating and need to be validated in larger, prospective, cardiovascular-focused studies.
Future studies can be focused on early detection and preventive management of subclinical CV disease associated with pazopanib and other agents within the VSPI class. This may include assessing the utility of more sensitive cardiac diagnostic modalities such as strain imaging for detection of subclinical LV dysfunction with novel VSPIs as has been shown with trastuzumab, anthracyclines, and taxanes [46]. Investigation of whether concurrent BB and/or ACEi/ARB use during pazopanib treatment imparts a cardioprotective effect as has been demonstrated in anthracycline-induced cardiomyopathy is another area that may warrant further investigation [47,48].