Cancer Risk in Pulmonary Hypertension Patients

Introduction

Pulmonary arterial hypertension (PAH; WHO group 1 pulmonary hypertension) is a rare but serious condition, characterized by elevated pulmonary artery pressure and contributing to increased morbidity and mortality.¹ Pulmonary hypertension (PH) may be idiopathic or secondary to several chronic diseases, including left-sided heart disease, chronic obstructive pulmonary disease, collagen vascular disease, and venous thromboembolism (VTE).¹

VTE is an established risk factor for chronic thromboembolic pulmonary hypertension (CTEPH; WHO group 4 PH),² although CTEPH occurs in patients with no history of symptomatic VTE in 19% to 63% of cases.³ Such patients may have suffered from asymptomatic VTE, but this remains unknown. An association between cancer and VTE has been recognized for at least 150 years.⁴ During recent decades, strong evidence has accumulated indicating that VTE may be a presenting symptom of cancer.^4,5

Similarly, exposure to carcinogenic substances, such as smoking and occupational hazards, is associated with group 3 and group 5 PH. Further, cancer itself - or its treatment - is known to cause pulmonary hypertension, particularly pulmonary tumor thrombotic microangiopathy and pulmonary veno-occlusive disease.¹ Hence it may be hypothesized that PH, like VTE, could be a marker of occult cancer. A small French cohort study of 346 patients with group 1, group 3, or group 4 PH found that 5-year cancer incidence approached 15%.⁶ Although this knowledge is very relevant for the management of PH patients, the association of cancer with PH compared to the general population is unknown. We therefore examined the association between a PH diagnosis and cancer risk in a cohort study using population-based data from the Danish health system.

Methods

The study was based on the cumulative Danish population of 7.6m persons during the 1995–2017 period. Danish residents have free access to tax-supported health care.⁷ We used the Danish National Patient Registry and the Danish Cancer Registry, which cover all Danish hospitals, to examine cancer incidence in 6335 patients with PH and without a previous cancer diagnosis. The civil registration number unique to every Danish citizen allowed linkage between the registries.⁷ Diseases in the registries are classified according to the International Classification of Diseases, Tenth Revision (ICD-10; ICD-8 codes were used before 1994). The study setting, registries, and variable definitions are described in detail in the Supplemental Material. We considered the date of first hospital admission or start of outpatient clinic follow-up to be the PH diagnosis date. We had access only to PH diagnoses based on ICD codes (Table S1). As a result, we lacked information on diagnostic criteria, severity, or subgroups of pulmonary hypertension, except for CTEPH, during part of the study period.

All the patients were followed from the PH diagnosis date until death or December 31, 2018, whichever came first. We calculated the expected number of cancer cases after the first hospital contact for PH, based on national incidence rates of first cancer diagnoses according to sex, age, and year of diagnosis in 1-year intervals. Multiplication of the number of person-years of follow-up by national incidence rates yielded the number of people with cancer that would be expected if patients with PH had the same risk of cancer as the general population. We used the standardized incidence ratio (SIR) – the ratio of observed number of cancers to expected number of cancers – to measure the association between PH and cancer, and calculated 95% confidence intervals (CIs). In the setting of a PH diagnosis, early cancer detection likely captures malignancies that would otherwise remain occult. For this reason, we examined the association with cancer according to arbitrarily chosen time periods since a PH diagnosis (≤1 year, >1 year). We performed a subgroup analysis for the period during which a specific ICD-10 code was available for CTEPH (2006–2017).

Treating death as a competing risk, we calculated the one-year risk of cancer following a PH diagnosis. Assuming that cancers detected during the first year of follow-up were present at the time of diagnosis, we computed the number of patients needing to be examined at the time of PH diagnosis in order to detect one excess cancer, as the reciprocal of the excess cancer risk.

We performed the analyses in SAS. The study was reported to the Danish Data Protection Agency (consent from patients is not required for Danish registry-based studies).

Results

We followed the 6335 patients for a median of 4.1 years. Table 1 provides descriptive data on the study cohort. During the first year of follow-up, 212 cancers were diagnosed. After the first year of follow-up, 796 cancer cases were diagnosed. The one-year risk was 3.3% and the one-year SIR was 1.96 (95% CI: 1.70–2.24). We found an elevated risk of several types of cancers (Figure 1). The cancer risk remained elevated even after one year of follow-up (SIR: 1.15 (95% CI: 1.08–1.24). The one-year SIR for patients with CTEPH was 1.41 (95% CI: 0.82–2.25). The number of patients with PH needing to be examined to detect one excess cancer was 52.

Table 1 Characteristics of the Study Cohort of Patients with Pulmonary Hypertension, Denmark, 1995–2017

Figure 1 Standardized Cancer Incidence Ratios for Patients with Pulmonary Hypertension. Abbreviations: O, observed number of cancer cases; E, expected number of cancer cases; SIR, standardized incidence ratio; CI, confidence interval.

Discussion

Cancer risk was clearly higher in patients with PH compared with the general population. We found a dichotomous pattern of cancer risk over time. In the first year after a PH diagnosis, cancer risk was particularly elevated. However, PH also was associated with long-term cancer risk. The same pattern is well established for the association between VTE and cancer.^4,5

Cancer may be associated with PH for several reasons. Heightened diagnostic efforts and the effects of occult cancer probably explain the short-term association, but we did not see a compensatory decreased incidence after the first year, characteristic for such a pattern. Other explanations for the persistent association are therefore required.

In summary, we identified an increased incidence of cancer in PH patients, although the absolute risks were low and the number needed to examine was high, limiting the clinical relevance of pursuing early cancer detection in these patients. Moreover, although our population-based cohort study was large and had virtually no loss to follow-up, our data lacked clinical detail and relied on ICD diagnoses. The high number needed to examine and these limitations prevent specific guidelines for cancer work-up in patients with different types of pulmonary hypertension and should be based on the clinical findings.