Background
Small-molecule receptor tyrosine kinase inhibitors (TKIs) targeting the VEGFR represent a mainstay systemic therapy for several advanced solid tumor malignancies.1 In particular, antiangiogenic VEGFR TKI therapies are a preferred systemic treatment approach for patients with advanced renal cell carcinoma (RCC) and thyroid cancer (TC), which represent the most common prescription indications. Indeed, prolonged VEGFR signaling inhibition with VEGFR TKIs is commonly used for several advanced TCs.2 In addition, 7 VEGFR TKI therapies have been approved by the FDA for the treatment of advanced RCC, either as monotherapy or in combination with immune checkpoint inhibitors (ICIs).3–5
Despite these important oncologic benefits, VEGFR TKI treatment results in clinically important cardiovascular (CV) toxicity. Numerous CV toxicities, including incident left ventricular dysfunction and hypertension, are class effects associated with VEGFR TKI therapies.6,7 Incident or worsening of preexisting hypertension represents the most common “on-target” CV toxicity, affecting up to 70% of patients.8,9 Although treatment-related hypertension while receiving VEGFR TKI therapy is reflective of therapeutic drug activity and is associated with improved cancer-related outcomes, such CV toxicities may also contribute to adverse CV outcomes and may result in treatment modifications, delays, or cessation of effective cancer therapy.10–12 These toxicities may be of mounting importance, given the improving long-term clinical outcomes recently observed with novel VEGFR TKI–based regimens across both TC and clear cell RCC malignancies.4,9
Yet, a comprehensive understanding of CV toxicity risk and adverse CV events in real-world patients who have cancer receiving VEGFR TKI therapies remains incomplete. First, although blood pressure (BP) management and CV risk mitigation guidelines in noncancer populations are informed by well-established patient-related risk factors and BP targets, much of the available data regarding clinical risk assessment and impact of VEGFR TKI–related CV toxicities in patients who have advanced cancer are derived from clinical trial populations, which may not be generalizable to patients cared for in everyday practice.13 Moreover, prior studies evaluating the impact of prevalent CV risk factors on the risk of hypertensive or other treatment-related CV events while receiving VEGFR TKI therapies are of limited sample size and have reported varied and inconsistent associations.8,14,15 Second, although adequate BP control in noncancer populations is known to significantly reduce morbidity and mortality from adverse CV events, the association between BP change and adverse CV events in patients who have advanced cancer receiving VEGFR TKI therapies is unknown.16,17 As a result, expert statements on CV risk assessment and BP control for patients receiving VEGFR TKI therapies have relied on standard risk assessment tools developed for noncancer populations.18 However, given the competing risks of cancer and noncancer adverse outcomes, the optimal BP target may be unique in cancer populations. Given these limitations, a number of unanswered questions remain. For example, what individual patient characteristics and treatment-related factors are most strongly associated with early BP changes? What are the associations between early “on-treatment” BP changes and subsequent adverse CV events? Are certain BP thresholds associated with improved clinical outcomes?
To address these current knowledge gaps and inform a real-world approach to VEGFR TKI hypertension and CV toxicity, we evaluated the key factors unique to an advanced cancer population that may inform the risks of early significant hypertension and adverse CV events, accounting for the competing risk of death. We first comprehensively characterized the baseline patient-, treatment-, and disease-related factors associated with early, acute treatment-related hypertension risk and with longer-term major adverse CV outcomes after initiation of VEGFR TKI therapy. In an exploratory analysis, we evaluated the association between on-treatment early BP elevations and risk of major adverse CV outcomes to gain insight into potential BP targets within this advanced cancer population.
Methods
Study Design and Data Source
We performed a multicenter retrospective cohort study of patients who have RCC and TC receiving cancer care at 3 large health systems: Penn Medicine and Barnes-Jewish Hospital/Washington University in St. Louis, both quaternary care academic medical centers, and the Geisinger Health System, a regional healthcare network serving central and northeastern Pennsylvania. Using the Penn Data Store, Clinical Desktop and Touchworks, and the Geisinger data warehouse, which are the enterprise data warehouses containing aggregated longitudinal clinical data from the electronic health record (EHR) at Penn Medicine, Barnes-Jewish Hospital/Washington University in St. Louis, and Geisinger Health System, respectively, we extracted information on patient demographics, hospital and office encounters, comorbidities, inpatient and outpatient medications, cancer therapies, procedures, and vital status using ICD-9-CM and ICD-10-CM diagnosis codes (supplemental eTable 1, available with this article at JNCCN.org). Available data related to VEGFR TKI administration, including dosages and prescription dates, were also extracted from the data warehouses. A full description of extracted data elements is provided in the Supplemental Methods (eAppendix 1).
The study cohort included patients with RCC or TC initiating first treatment with VEGFR TKI therapy, identified via an outpatient or inpatient prescription, with a treatment start date between January 1, 2008, and May 31, 2020, and a corresponding clinical provider visit. To identify patients with a new exposure to VEGFR TKI therapy followed longitudinally at the study site, an observable period of at least 90 days without VEGFR TKI exposure and a minimum duration of 30 days between the VEGFR TKI start date and the last observed clinical provider visit date were required. The study protocol was reviewed and approved by the Institutional Review Boards at each participating site, and informed consent was not required.
Covariate and Outcomes Definitions
Prevalent hypertension was defined as having either ICD-9-CM or ICD-10-CM diagnosis codes recorded on at least one encounter or the prescription of an antihypertensive medication during the 90 days before VEGFR TKI initiation (supplemental eTable 1). Hypertensive medications included β-blockers, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, diuretics, calcium channel blockers, and nitrates, as well as unclassified antihypertensives such as hydralazine. Other baseline CV risk factors (hyperlipidemia, diabetes mellitus) and CV disease (heart failure [HF], coronary artery disease, peripheral artery disease, and cerebrovascular accident [CVA]) were defined by ICD-9-CM or ICD-10-CM diagnosis codes present during at least one encounter before VEGFR TKI initiation. Baseline BP was calculated as the mean of recorded values in the 90 days before VEGFR TKI initiation. Major adverse cardiovascular events (MACE) after VEGFR TKI initiation represented a composite outcome of CV-related medical encounters, including HF (occurring in at least 1 inpatient encounter or 2 outpatient encounters at least 30 days apart), acute coronary syndrome, myocardial infarction, coronary revascularization (occurring as primary or secondary inpatient diagnostic codes), or CVA (occurring as the primary inpatient diagnostic code). For patients with a history of baseline HF, incident HF was only included if occurring as a primary inpatient diagnosis code, indicating hospitalization for exacerbation of prevalent HF. All CV diagnoses were ascertained by the clinical provider at the time of the health system encounter. The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation.19 Patients receiving concomitant ICI therapy with VEGFR TKI initiation were defined as having the presence of an ICI treatment order within 60 days of the initial VEGFR TKI prescription.
Statistical Analyses
We first sought to characterize prevalent hypertension and CV comorbidity at the time of initiation of VEGFR TKI therapy (baseline). Baseline patient characteristics were summarized using proportions for categorical variables and medians with 25th and 75th percentiles (Q1–Q3) for continuous variables. The chi-square test or ANOVA was used for statistical comparisons across groups (cancer type, race).
We then sought to evaluate the associations between baseline patient- and treatment-related variables and the outcomes of early significant hypertension, MACE, and overall survival (OS) after VEGFR TKI initiation. First, to evaluate the associations between baseline characteristics with the outcome of early hypertension after VEGFR TKI initiation, we evaluated BP measures on a continuous scale. Multivariable linear regression models were used to evaluate the association between baseline characteristics and early change in systolic BP (SBP) or diastolic BP (DBP) from baseline. We evaluated this BP outcome within 6 weeks after VEGFR TKI initiation, given the known significant early effects of VEGFR TKIs on BP and to minimize confounding by CV or oncologic therapy. In the event of multiple BPs obtained for a patient within a single day, the mean BP within a day was used. Independent variables in the multivariable models were selected a priori based on clinical judgment and included age, race, sex, clinical treatment site, baseline SBP, VEGFR TKI type, diabetes mellitus, hyperlipidemia, cancer type, and baseline use of any antihypertensive medication. Interaction analyses for prespecified, hypothesized effect modifiers, including cancer type, race, and history of hypertension, were also conducted.
The incidence of composite MACE and its association with baseline patient- and treatment-related characteristics was evaluated using a competing risks framework to account for the risk of death in an advanced cancer population. To evaluate the prognostic factors associated with MACE incidence, the distribution of time to first MACE after VEGFR TKI treatment initiation was evaluated using cumulative incidence functions (CIFs), and Gray’s test was used to compare CIFs among subgroups defined by baseline factors. Subdistribution hazard ratios (HRs), the ratios of the instantaneous risks of having MACE between subgroups defined by baseline factors, were estimated. Covariates included in an adjusted MACE competing risk model were selected a priori based on clinical judgment and included age, race, sex, treatment site, cancer type, diabetes mellitus, baseline SBP >140 mm Hg, hyperlipidemia, and history of CV disease.
In an exploratory analysis, the association between early on-treatment BP change during the initial treatment period and incident MACE was evaluated using a competing risk framework accounting for the risk of death. Early on-treatment BP change was defined as the difference between mean baseline SBP (or DBP) and the peak SBP (or DBP) within 4 weeks of VEGFR TKI initiation. On-treatment early SBP or DBP changes were categorized as an increase of <10 mm Hg (reference), 10–20 mm Hg, and ≥20 mm Hg. CIFs with estimates of subdistribution HR were used to assess the association between early on-treatment BP changes with incident adverse CV outcomes (the first of MACE or death), adjusted for age, race, sex, cancer type, baseline CV disease, and baseline use of any antihypertensive medication. For these time-to-event analyses, patients were followed until the first of MACE or death, or they were censored at 2 years after VEGFR TKI initiation to minimize confounding.
Finally, multivariable Cox regression models were used to define the predictors of all-cause survival. Clinical factors of interest that were determined a priori included age, race, sex, treatment site, baseline eGFR, history of prior nephrectomy, hypertension, CV disease, body mass index (BMI; categorical >30 kg/m2), and cancer type. For the time-to-event MACE and OS analyses, patients were followed from VEGFR-TKI initiation to the first event or death, or they were censored at the date of the last follow-up visit.
Model regression coefficient estimates with 95% confidence intervals are presented for linear regression models, odds ratios with 95% confidence intervals are presented for logistic regression models, and HRs with 95% confidence intervals are presented for subdistribution and Cox proportional hazards models. A 2-sided alpha level of 0.05 was used to assess statistical significance. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc.).
Results
Patient Population and Baseline Hypertension and CV Morbidity
A total of 987 eligible patients, including 793 (80.3%) with RCC and 194 (19.7%) with TC, were included in the study cohort (Table 1), with baseline characteristics by treatment site shown in supplemental eTable 2. Among patients with RCC, the median age of the study cohort at baseline was 63 years (Q1–Q3, 56–71 years), 85.2% were White, and 8.4% were Black. The most common VEGFR TKIs initiated in patients with RCC were pazopanib (40.2%) and sunitinib (37.7%), whereas the most common VEGFR TKIs newly initiated in patients with TC were sorafenib (40.2%) and lenvatinib (24.7%). Among patients with RCC, 64.6% had prior nephrectomy, and 59.9% had chronic kidney disease (median eGFR, 56.8 [Q1–Q3, 42.0–76.8] mL/min/1.73m2). The prevalence of hypertension and CV disease at baseline was high: Among patients with RCC and TC, respectively, 61.5% and 53.6% had hypertension, 19.8% and 12.4% had coronary artery disease, and 8.8% and 6.2% had HF. One-third of patients (n=329) had an average SBP >140 mm Hg (median, 148 [range, 139–206] mm Hg) during the 90 days before VEGFR TKI initiation. Differences were noted in the baseline CV risk profile according to race (supplemental eTable 3), with a higher proportion of Black patients with RCC having hypertension (80.6% vs 60.5%; P<.01 overall across race), CVA/stroke (16.4% vs 8.3%; P=.04), and CKD (77.6% vs 59.3%; P<.01), and a numerically greater proportion having prevalent HF (14.9% vs 8.6%; P=.10) and diabetes (27.3% vs 24.6%; P=.07). Although the baseline SBP and DBP were similar between races, a greater proportion of Black patients than White patients were receiving antihypertensive medications at baseline (71.6% vs 47.3%; P<.01).
Baseline Patient Characteristics (N=987)
Incidence and Predictors of Early Hypertension After VEGFR TKI Initiation
We then characterized the incidence and baseline predictors of early hypertension during the first 6 weeks after the initiation of VEGFR TKI therapy. The median number of BP measures per patient during the first 6 weeks was 2 (Q1–Q3, 2–4). One BP measure was recorded in 22.1% of patients, and 47.8% of patients had ≥3 BP measures during the 6-week period. Totals of 530 (53.7%) and 172 (17.4%) patients had at least one recorded SBP >140 mm Hg or >160 mm Hg, respectively, during this initial treatment period, including 185 (46.8%) and 46 (11.7%) without baseline hypertension. Similarly, totals of 212 (21.5%) and 42 (4.3%) patients had at least one recorded DBP >90 mm Hg or >100 mm Hg, respectively, including 84 (21.3%) and 14 (3.5%) without baseline hypertension. Among all patients, totals of 276 (28.0%) and 370 (37.5%) experienced an increase in SBP ≥20 mm Hg or DBP ≥10 mm Hg, respectively, from baseline.
The associations between baseline clinical variables and early systolic or diastolic BP changes are summarized in Table 2. In multivariable analyses, elevated baseline SBP was strongly associated with a significant mean BP increase after VEGFR TKI initiation. Similarly, Black race was associated with an increase in mean SBP from baseline on the order of 3.2 mm Hg (95% CI, 0.4–6.1 mm Hg). Notably, no significant associations were observed between sex, baseline diabetes mellitus, cancer type, or specific VEGFR TKI therapy with significant early BP changes. On evaluation of hypothesized effect modifiers, no significant interaction was observed between baseline hypertension and cancer type for early BP change outcomes. However, in multivariable analysis adjusted for all variables noted in Table 2, including baseline antihypertensive use, a significant interaction between baseline hypertension and race was present, with Black patients without baseline hypertension having a greater change in SBP from baseline after VEGFR TKI initiation (mean SBP change, 14.4 mm Hg; 95% CI, 8.14–20.7 mm Hg) than Black patients with baseline hypertension (mean SBP change, 4.4 mm Hg; 95% CI, 0.9–7.8 mm Hg).
Association Between Baseline Clinical Variables and Early Change in SBP or DBP After VEGFR TKI Initiation
Incidence and Predictors of First MACE After VEGFR TKI Initiation
We next evaluated whether baseline patient-, treatment-, and disease-related factors were associated with increased risk for major adverse CV outcomes after initiation of VEGFR TKI therapy. A total of 147 MACE occurred over a median follow-up of 19 months and a maximum follow-up of 158 months, which included first occurrences of 71 incident HF events (48.3% of MACE outcomes), 49 incident CVA events (33.3% of MACE), and 27 acute coronary events or coronary revascularizations (18.4% of MACE). The cumulative incidences of the first MACE at 1 and 2 years were 6.9% and 10.0%, respectively, and 68 (46.3%) of 147 observed events occurred within 1 year of VEGFR TKI initiation. The median time to a MACE was 14 months (Q1–Q3, 4–31 months). The associations between baseline clinical variables and incident MACE after VEGFR TKI initiation are summarized in Table 3. Using a competing risks modeling framework with Fine and Gray analysis, baseline SBP >140 mm Hg (subdistribution HR, 1.41; 95% CI, 1.00–1.98) and history of CV disease (subdistribution HR, 1.62; 95% CI, 1.15–2.28) were each associated with MACE outcomes in univariable analyses (Figure 1A, B). However, no significant associations between CV disease, diabetes, hyperlipidemia, SBP >140 mm Hg, or male sex and MACE outcomes were observed in multivariable analysis accounting for the competing risk of death (Table 3). Similarly, there were no significant associations between CV disease or CV risk factors and MACE outcomes in stratified analyses by cancer type (supplemental eTables 4 and 5).
Association Between Baseline Clinical Variables and Major Adverse Cardiovascular Events After VEGFR TKI Initiation
Cumulative incidence of major adverse CV events after VEGFR tyrosine kinase inhibitor initiation, stratified by baseline (A) SBP and (B) CV disease. (A) Unadjusted cumulative incidence function of MACE, by baseline SBP >140 mm Hg (subdistribution HR, 1.41; 95% CI, 1.00–1.98). (B) Unadjusted cumulative incidence function of MACE, by baseline CV disease (subdistribution HR, 1.62; 95% CI, 1.15–2.28).
Abbreviations: CV, cardiovascular; HF, heart failure; HR, hazard ratio; MACE, major adverse cardiovascular events; PAD, peripheral artery disease; SBP, systolic blood pressure; TIA, transient ischemic attack.
Citation: Journal of the National Comprehensive Cancer Network 21, 10; 10.6004/jnccn.2023.7047
Association Between On-Treatment BP Change and Adverse CV Events
In an exploratory analysis, we next sought to evaluate the association between early changes in on-treatment BP and subsequent risk of MACE to inform potential optimal BP targets for mitigating CV risk in patients initiating VEGFR TKI therapy. A total of 304 patients (30.8%) experienced a mean SBP >140 mm Hg, and 321 patients (32.5%) experienced a mean DBP >80 mm Hg within 4 weeks of VEGFR TKI initiation. Subdistribution HR competing risk analysis evaluating the association between early on-treatment BP change during the initial 4 weeks of VEGFR TKI therapy and MACE outcomes is summarized in Table 4. Early significant elevations of SBP or DBP during the first 4 weeks after VEGFR TKI initiation were not significantly associated with adverse CV outcomes within 2 years (subdistribution HRs for on-treatment elevations of SBP >20 mm Hg and DBP >20 mm Hg: 0.99; 95% CI, 0.60–1.65 and 1.55; 95% CI, 0.76–3.18, respectively).
Association Between Initial On-Treatment SBP or DBP Control and Major Adverse Cardiovascular Events During First Year After VEGFR TKI Initiation
Predictors of OS After VEGFR TKI Initiation
A total of 592 patients (60.0%) died over a maximum follow-up of 158 months. Over a median follow-up of 19 months after VEGFR TKI initiation, the median time to death was 13 months (Q1–Q3, 5.5–27 months). The 1-year and 2-year all-cause mortality rates after VEGFR TKI initiation were 26.7% and 42.5%, respectively. Table 5 demonstrates the associations between baseline clinical variables and overall survival after VEGFR TKI initiation. Within this cohort, patients with RCC had inferior survival compared with those with TC (HR, 1.61; 95% CI, 1.22–2.13; P<.001). Consistent with established prognostic factors for advanced RCC, a history of prior nephrectomy, which typically indicates initial presentation with clinically localized disease or candidacy for cytoreductive nephrectomy, was associated with improved survival after VEGFR TKI initiation (HR, 0.64; 95% CI, 0.53–0.79; P<.001).20 Black race was associated with inferior survival (HR, 1.40; 95% CI, 1.04–1.89; P=.03). However, known CV risk factors (eg, hypertension, BMI ≥30 kg/m2) and prevalent CV disease before VEGFR TKI initiation were not significantly associated with OS outcomes.
Association Between Baseline Clinical Variables and Overall Survival After VEGFR TKI Initiation
Discussion
In this multicenter retrospective cohort study of patients receiving cancer care across 3 large US health systems, we first evaluated the baseline prevalence and incidence of early hypertension and MACE outcomes after VEGFR TKI therapy initiation for the most common prescription indications, advanced RCC or TC malignancies. In this real-world nonclinical trial population, the baseline prevalence of CV disease and traditional CV risk factors, in particular hypertension, was notably high, with more than half of patients diagnosed with hypertension and prescribed antihypertensive agents. In addition, adverse CV events, including HF and CVA events, were common (occurring in 14.9% of patients overall) and frequently occurred early (46.3% of events occurred within 1 year of VEGFR TKI initiation). When evaluating the association between baseline patient-, treatment-, and disease-related factors and early hypertensive and adverse CV outcomes, a key finding from our analyses was that baseline hypertension and Black race were the primary clinical factors associated with greater risk of early BP increases while receiving VEGFR TKI therapies. These findings lend insight into the identification of high-risk subgroups who may be more likely to experience significant acute hypertension with the initiation of VEGFR TKIs.
An improved understanding of the clinical impact of this early significant hypertension and other traditional CV risk factors on adverse CV outcomes, particularly within an advanced cancer population at high risk for cancer-related mortality, is key to informing the optimal management of VEGFR TKI–related hypertension. Patients who have advanced cancer, in particular those with RCC, are at increased risk for competing CV morbidity and mortality due to the shared risk factors between cancer and CV disease and the known cancer therapy toxicity profiles.6 In our analyses, elevated baseline SBP >140 mm Hg and CV disease were associated with MACE in univariable analysis, in keeping with previous published studies of VEGFR TKI therapies indicating a prognostic association between baseline hypertension and prevalent CV events with MACE.15 However, in our study cohort, no baseline clinical factors, including traditional CV risk factors, were associated with MACE after VEGFR TKI initiation in adjusted analyses. Although a MACE incidence of 10% at 2 years in our cohort suggests that CV morbidity in this population is important, disease-related risk in this advanced cancer population remained a significant competing risk, with a median survival of 13 months, most likely secondary to progressive malignancy. Similarly, because hypertension is a known risk factor for adverse CV outcomes, we had hypothesized that treatment-related early hypertensive toxicity while receiving VEGFR TKI therapy would contribute to VEGFR TKI–related MACE, including HF and CVA.21 However, in our study, early on-treatment BP elevations were not associated with MACE during the first 2 years after VEGFR TKI initiation. As such, although prior randomized clinical trials have demonstrated the benefits of intensive BP control (to a target SBP <120 mm Hg) for decreasing MACE and all-cause mortality in noncancer populations, the appropriate BP targets in this unique cancer population remain an important unanswered question.22 In particular, although our analyses indicated that early changes in BP (within 4 weeks of VEGFR TKI initiation) were not associated with MACE, the association between more sustained BP elevations while receiving VEGFR TKI therapy and adverse CV outcomes in this population remains uncertain.
Taken together, these findings may inform several important aspects of real-world clinical management of early hypertensive toxicity in patients who have advanced cancer and are initiating VEGFR TKIs. Because hypertension is a well-established toxicity of VEGFR TKI therapies, eligibility criteria for clinical trials and resulting regulatory product labels encourage adequate BP control before treatment initiation.23 However, the nontrial population included in this study cohort reflects the variable baseline BP control common to patients who have RCC and TC and are initiating VEGFR TKIs in routine clinical practice, in which one-third of patients initiating VEGFR TKI therapy had a mean SBP >140 mm Hg during the 3 months before treatment initiation. Moreover, our findings indicate that elevated baseline SBP is the strongest factor associated with the development of clinically significant acute changes in BP after VEGFR TKI initiation, even when accounting for the baseline use of antihypertensive medications. These findings underscore the current undermanagement of this modifiable risk factor and the relevance of adequate BP control before VEGFR TKI initiation to mitigate early on-treatment hypertensive toxicity.8,14 These findings provide evidence to support the cardio-oncology consensus recommendations from the European Society of Cardiology (ESC), which emphasize recognition and management of hypertension before therapy initiation.24 In addition, although BP elevations while receiving VEGFR TKI therapy have been associated with improved cancer outcomes, early on-treatment significant BP elevations may often contribute to VEGFR TKI cessation, dose interruption, or reduction (given the known association between poor BP control and adverse CV outcomes in noncancer populations). However, our findings indicate that such acute BP elevations were not associated with MACE and may emphasize that priority should be given to continuing VEGFR TKI therapy in the early treatment period to maximize cancer treatment, with parallel efforts to carefully manage and control BP, consistent with the concept of permissive cardiotoxicity.25 These findings similarly support recent ESC recommendations, which state a “goal to continue VEGF inhibitor treatment for as long as possible with initiation or optimization of CV treatment if indicated.”24 However, the potential longer-term CV effects of such permissive hypertension during more sustained BP elevations while receiving VEGFR TKI therapy require further evaluation.
An advanced cancer population treated with palliative-intent systemic therapies represents a unique setting in which to assess and manage comorbid CV risks. Importantly, the highly variable clinical behavior of advanced RCC and TC, with some favorable-risk patients experiencing prolonged survival outcomes, is well established.26 Furthermore, contemporary treatment regimens combining VEGFR TKIs with other cardiotoxic agents, such as ICI therapy, have significantly improved the prognosis in this population, now with a median survival approaching 4 years.27 Consequently, this increased survival may also result in greater clinical significance of CV comorbidities in the longer term for some patients. As such, our findings overall indicate that traditional CV risk assessment tools may not broadly apply within this cancer population and that cancer-specific CV risk assessments, incorporating both cancer-related prognostic factors and CV-related patient factors (in particular elevated baseline SBP >140 mm Hg and prevalent CV disease) may be necessary to optimally identify patients at highest risk of MACE after VEGFR TKI therapy.
An additional notable finding from our study was the association between self-identified Black race and both hypertension and worse survival outcomes. Black race was associated with inferior OS, as well as a greater change from baseline in BP among nonhypertensive patients, after VEGFR TKI initiation in adjusted analyses. Similar inferior clinical outcomes for Black patients relative to White patients receiving VEGFR TKI therapies have been reported.28 Although the reasons for this difference remain unclear, the higher proportion of baseline comorbidities among Black patients, as well as potential disparities in care, including less-aggressive antihypertensive management or undertreated baseline hypertension, may contribute to observed differences in hypertensive outcomes while receiving VEGFR TKI therapy and warrant further investigation. These findings again highlight patient populations that may be particularly vulnerable and the need for a targeted, multidisciplinary approach to improve care.
Our study had limitations. First, although representing a large multi-institutional cohort of patients who have cancer, potential limitations in power may have resulted in a lack of a detectable association between baseline CV risk factors (eg, diabetes mellitus, hyperlipidemia, elevated BMI) and adverse hypertensive or MACE outcomes. Second, although the study cohort was restricted to patients receiving longitudinal oncologic care at one of the study institutions, CV risk factors, diagnoses, and medications initiated at outside health systems may not have been captured, resulting in a risk of misclassification. Similarly, the use of ICD codes as documented in the EHR medical history may have resulted in misclassification in baseline and/or incident MACE outcomes. For example, although incident HF was included as an MACE only if listed as a primary inpatient diagnosis code among patients with baseline HF, this may have overestimated the true incidence of HF after VEGFR TKI initiation. We also did not have information on the cause of death. Third, we were unable to account for the potential effects of antihypertensive or VEGFR TKI dose modifications during the initial 6 weeks after TKI initiation. Therefore, although our study findings indicate that patients with early increases in BP may benefit from VEGFR TKI continuation, future studies are needed to evaluate the effect of early TKI dose modifications on longer-term BP changes and/or MACE, because we did not have detailed longitudinal BP data over an extended follow-up. Fourth, although ESC guidelines recommend ambulatory/home BP monitoring while receiving VEGFR TKI therapy to minimize “white coat” hypertension and variations in measurement technique, our study relied on EHR-derived BP measures obtained in a clinic setting.24 Finally, the median survival time in our cohort was 13 months, likely related to the low use of ICIs with VEGFR TKIs given the study time period (2008–2020) and evolving practice patterns. With more recent significant improvements in survival outcomes, the prospect for competing CV risk acquires increasing clinical relevance. Strengths of this analysis include the large study population across multiple regional health systems, different cancer types, and diverse VEGFR TKI exposures. In addition, the included health systems represent regional health networks with integrated EHRs across medical disciplines and community/academic treatment sites.
Conclusions
Our study demonstrates that hypertension and CV morbidities are highly prevalent among patients initiating VEGFR TKI therapies and that baseline hypertension and Black race represent the primary clinical factors associated with VEGFR TKI–related early significant hypertension, suggesting that there is a need for greater awareness in the management and care of these subgroups. Although adverse CV events are common, occur early in this advanced cancer population, and may impart significant medical and quality-of-life implications with steadily improving long-term cancer outcomes, our findings suggest that early BP elevations are not associated with MACE in this population and that unique cancer-specific CV risk assessment tools are needed for patients treated with VEGFR TKIs. Further evaluations of the level of optimal BP control and its association with cancer, CV, and quality of life/symptom outcomes are warranted. A cluster randomized trial evaluating the feasibility of achieving intensive versus standard BP control in patients initiating contemporary VEGFR TKI–based regimens (with or without ICI therapy) is currently ongoing (ClinicalTrials.gov identifier: NCT04467021).
References
- 1.↑
Li W, Croce K, Steensma DP, et al. Vascular and metabolic implications of novel targeted cancer therapies: focus on kinase inhibitors. J Am Coll Cardiol 2015;66:1160–1178.
- 2.↑
Brose MS. Sequencing of tyrosine kinase inhibitors in progressive differentiated thyroid cancer. Clin Adv Hematol Oncol 2016;14(Suppl 9):7–12.
- 3.↑
Choueiri TK, Powles T, Burotto M, et al. Nivolumab plus cabozantinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med 2021;384:829–841.
- 4.↑
Motzer R, Alekseev B, Rha SY, et al. Lenvatinib plus pembrolizumab or everolimus for advanced renal cell carcinoma. N Engl J Med 2021;384:1289–1300.
- 5.↑
Castro DV, Malhotra J, Meza L, et al. How to treat renal cell carcinoma: the current treatment landscape and cardiovascular toxicities. JACC CardioOncol 2022;4:271–275.
- 6.↑
Narayan V, Ky B. Common cardiovascular complications of cancer therapy: epidemiology, risk prediction, and prevention. Annu Rev Med 2018;69:97–111.
- 7.↑
Chen DY, Liu JR, Tseng CN, et al. Major adverse cardiovascular events in patients with renal cell carcinoma treated with targeted therapies. JACC CardioOncol 2022;4:223–234.
- 8.↑
Hall PS, Harshman LC, Srinivas S, et al. The frequency and severity of cardiovascular toxicity from targeted therapy in advanced renal cell carcinoma patients. JACC Heart Fail 2013;1:72–78.
- 9.↑
Schlumberger M, Tahara M, Wirth LJ, et al. Lenvatinib versus placebo in radioiodine-refractory thyroid cancer. N Engl J Med 2015;372:621–630.
- 10.↑
Rini BI, Cohen DP, Lu DR, et al. Hypertension as a biomarker of efficacy in patients with metastatic renal cell carcinoma treated with sunitinib. J Natl Cancer Inst 2011;103:763–773.
- 11.↑
Rini BI, Schiller JH, Fruehauf JP, et al. Diastolic blood pressure as a biomarker of axitinib efficacy in solid tumors. Clin Cancer Res 2011;17:3841–3849.
- 12.↑
Wirth LJ, Tahara M, Robinson B, et al. Treatment-emergent hypertension and efficacy in the phase 3 study of (E7080) lenvatinib in differentiated cancer of the thyroid (SELECT). Cancer 2018;124:2365–2372.
- 13.↑
James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507–520.
- 14.↑
Hamnvik OP, Choueiri TK, Turchin A, et al. Clinical risk factors for the development of hypertension in patients treated with inhibitors of the VEGF signaling pathway. Cancer 2015;121:311–319.
- 15.↑
Vallerio P, Orenti A, Tosi F, et al. Major adverse cardiovascular events associated with VEGF-targeted anticancer tyrosine kinase inhibitors: a real-life study and proposed algorithm for proactive management. ESMO Open 2022;7:100338.
- 16.↑
Ogden LG, He J, Lydick E, et al. Long-term absolute benefit of lowering blood pressure in hypertensive patients according to the JNC VI risk stratification. Hypertension 2000;35:539–543.
- 17.↑
Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002;360:1903–1913.
- 18.↑
Maitland ML, Bakris GL, Black HR, et al. Initial assessment, surveillance, and management of blood pressure in patients receiving vascular endothelial growth factor signaling pathway inhibitors. J Natl Cancer Inst 2010;102:596–604.
- 19.↑
Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–612.
- 20.↑
Heng DY, Xie W, Regan MM, et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted agents: results from a large, multicenter study. J Clin Oncol 2009;27:5794–5799.
- 21.↑
Catino AB, Hubbard RA, Chirinos JA, et al. Longitudinal assessment of vascular function with sunitinib in patients with metastatic renal cell carcinoma. Circ Heart Fail 2018;11:e004408.
- 22.↑
Wright JT Jr, Williamson JD, Whelton PK, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med 2015;373:2103–2116.
- 23.↑
Prescribing information for sunitinib malate capsules, oral. 2011. Accessed May 1, 2023. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2011/021938s13s17s18lbl.pdf
- 24.↑
Lyon AR, López-Fernández T, Couch LS, et al. 2022 ESC guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS). Eur Heart J 2022;43:4229–4361.
- 25.↑
Porter C, Azam TU, Mohananey D, et al. Permissive cardiotoxicity: the clinical crucible of cardio-oncology. JACC CardioOncol 2022;4:302–312.
- 26.↑
Ko JJ, Xie W, Kroeger N, et al. The International Metastatic Renal Cell Carcinoma Database Consortium model as a prognostic tool in patients with metastatic renal cell carcinoma previously treated with first-line targeted therapy: a population-based study. Lancet Oncol 2015;16:293–300.
- 27.↑
Rini BI, Plimack ER, Stus V, et al. Pembrolizumab (pembro) plus axitinib (axi) versus sunitinib as first-line therapy for advanced clear cell renal cell carcinoma (ccRCC): results from 42-month follow-up of KEYNOTE-426. J Clin Oncol 2021;39(Suppl):Abstract 4500.
- 28.↑
Bossé D, Xie W, Lin X, et al. Outcomes in Black and white patients with metastatic renal cell carcinoma treated with first-line tyrosine kinase inhibitors: insights from two large cohorts. JCO Glob Oncol 2020;6:293–306.
