Bronchiectasis and the risk of cancer: a nationwide retrospective cohort study.

Bronchiectasis and the risk of cancer: a

nationwide

retrospective cohort study

W.-S. Chung,

C.-L. Lin,

C.-L. Lin,

C.-H. Kao

Introduction

Bronchiectasis, a lung disease characterized by localized damage and irreversible dilatation of the bronchial tree, is caused by recurrent inflammation or infection of the airway. Inflammation is a crucial

innate immune response that disturbs tissue homeostasis, and chronic inflammation may play a decisive

role in tumour development and affect immune surveillance (1,2). Chronic obstructive pulmonary disease

(COPD) is a systemic inflammatory disorder that increases the risk of lung cancer (3,4). Previous studies have indicated that infection is

related to cancer development. For example, persistent infection of carcinogenic human papillomavirus

can cause cervical cancer (5). The Helicobacter pylori infection without eradication is associated with gastric cancer and mucosa-associated lymphoid tissue lymphoma (6). Infections with hepatitis B or C viruses increase the risk of hepatocellular carcinoma (7,8), and infections with Schistosoma or Bacteroides species are linked to bladder and colon cancer,

respectively (9,10). Patients infected with Mycobaterium tuberculosis have an increased risk of lung

cancer (11). Infection can activate multiple oncogenic pathways and facilitate the development of

tumours (12).

Patients with bronchiectasis have been reported to exhibit increased systemic inflammation (13). However, studies investigating the relationship between

conducted a nationwide population-based cohort study to explore whether bronchiectasis increases the risk of cancer.

Method

Data source

In this study, we used data from the National Health Insurance Research Database (NHIRD) in Taiwan. The Taiwan National Health Insurance (NHI) program was established on 1 March, 1995, providing

comprehensive medical care coverage for all residents nationwide. By 2009, approximately 99% of the 23.74 million people in Taiwan were enrolled in the program (14). The data from NHI claims include comprehensive demographic data, the dates of clinical visits, diagnostic codes, and details on prescriptions, examinations and procedures. This

information is entered into the NHIRD and maintained by the Bureau of National Health Insurance

(BNHI). In the NHI program, insurants who have

certain major diseases, such as malignancies, transplants or autoimmune diseases, can apply for a catastrophic illness certificate. Registering a patient with

a catastrophic illness requires a physician diagnosis and confirmatory pathological results or other supporting medical information; these documents are

formally reviewed by the BNHI. To ensure that personal information is protected, patient identification

is scrambled by the BNHI before it is released for public use. The patient diagnoses are coded according to the International Classification of Diseases,

Ninth Revision, Clinical Modification (ICD-9-CM). Numerous studies have indicated a high accuracy and validity of the ICD-9-CM diagnosis in the NHIRD (15,16). This study was approved from review by the Institutional Review Board of China Medical University in Central Taiwan (CMU-REC-101-012).

Using inpatient claims data, we identified patients 20 years of age and older who had been diagnosed with bronchiectasis (ICD-9-CM code 494) between 1 January 1998 and 31 December 2010. The first admission for bronchiectasis served as the index date. We extracted 53,755 bronchiectasis patients for whom age and sex information was complete and who did not have a history of cancer (ICD-9-CM codes 140–208) before the index date to be the bronchiectasis cohort. Four non-bronchiectasis control patients for each bronchiectasis patient were frequency-matched with the bronchiectasis cohort according to age group with a 5-year interval, sex, and the year of index date. Control

patients with a history of cancer before the index date or for whom age or sex information was incomplete were excluded and replaced with another qualified control patient. Finally, a total of 53,755 patients with bronchiectasis and 215,020 patients without bronchiectasis were included in this study. Taiwan launched a NHI in 1995, operated by a single-buyer, the government. Medical

reimbursement specialists and peer review should scrutinize all insurance claims. The diagnoses of bronchiectasis were based on the ICD-9 codes

which were judged and determined by related specialists and physicians according to the standard

clinical criteria and CT imaging revealed tree-inbud abnormalities, dilated bronchi and cysts with

defined borders. Therefore, the diagnoses and codes for bronchiectasis used in this cohort study should be correct and reliable.

Main outcome and comorbidities

Cancer patients were identified from the Registry for Catastrophic Illness Patient Database (RCIPD) in Taiwan. Study patients were observed from the index date to the date of cancer diagnosis, withdrawal from the insurance program, censoring

because of death, or the end date in the database (31 December 2011). Comorbidities before the index date obtained from the inpatient claims data were diabetes (ICD-9-CM code 250), hypertension 9-CM codes 401–405), hyperlipidaemia (ICD-9-CM code 272), and COPD (ICD-(ICD-9-CM codes 491, 492, 496).

Statistical analysis

We analyzed the data, comparing the distributions of age, sex and baseline comorbidities between the bronchiectasis cohort and the non-bronchiectasis cohort. The v2 test was used for categorical variables,

and the t test was used for continuous variables. The incidence densities for cancer were

assessed for each cohort according to sex, age and comorbidity. The relative incidence rate ratio (IRR) of patients with bronchiectasis who developed cancer compared with patients without bronchiectasis who developed cancer was analyzed using Poisson

regression. Multivariable Cox proportional hazardregression analyses were employed to calculate the

adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for the cancer risk associated with bronchiectasis. The multivariable models were simultaneously adjusted according to age, sex and

comorbidities, namely diabetes, hypertension, hyperlipidaemia and COPD. All data processing and

statistical analyses were performed using the SAS _{software version 9.2 (SAS Institute, Inc.,}

Cary,

NC). A two-tailed p value of < 0.05 was considered statistically significant.

Results

Demographic characteristics and comorbidity

between bronchiectasis patients and the

comparison cohort

Table 1 shows the distribution of demographic characteristics and baseline comorbidity statuses for the two cohorts. The mean age of patients with

bronchiectasis was 68.1 years [standard deviation (SD) 15.2], and the mean age of patients without bronchiectasis was 67.2 years (SD 15.3). More than half of the patients with bronchiectasis were aged ≤ 65 years, and 53.1% of the cohort consisted of men. Comorbidities, namely diabetes (19.2% vs. 8.81%), hypertension (35.8% vs. 17.0%),

hyperlipidaemia (6.57% vs. 3.58%), and COPD (35.0% vs. 4.99%) were more prevalent in the bronchiectasis cohort than in the non-bronchiectasis cohort. The mean lengths of follow-up were

4.76 years (SD = 4.73 years) and 6.36 years (SD = 6.34 years) for the bronchiectasis cohort and the non-bronchiectasis cohort, respectively (data not shown).

Incidence and aHRs of cancer stratified

according to sex, age and comorbidity in the

bronchiectasis and non-bronchiectasis cohorts

In total, 4345 cancer patients were identified in the bronchiectasis cohort, representing an incidence rate of 17.0 per 1000 person-years, and 16,668 cancer patients were identified in the non-bronchiectasis cohort, constituting an incidence rate of 12.2 per 1000 person-years. Accordingly, the IRR was 1.39 (95% CI =1.36–1.43), and the aHR was 1.46 (95% CI = 1.41–1.52) (Table 2). The incidence of cancer was slightly higher in men than in women and

increased with age in both cohorts. The risk of cancer for bronchiectasis patients was significantly higher than that for the comparison cohort in both sexes after adjusting for covariates (aHR = 1.38, 95% CI = 1.30–1.46 in women and aHR = 1.52, 95%

CI = 1.45–1.60 in men). The age-specific bronchiectasis cohort to non-bronchiectasis cohort aHRs of cancer were significant for all age groups (aHR = 1.75, 95% CI = 1.51–2.03 in the 20–49 year-old age group; aHR = 1.66, 95% CI = 1.53–1.80 in the 50–64 yearold age group; aHR = 1.43, 95% CI = 1.34–1.53 in

the 65–74 year-old age group; aHR = 1.33, 95% CI = 1.25–1.41 in the ≥ 75 year-old age group). In addition, the incidence of cancer increased with the

number of comorbidities in both cohorts.

Incidence, IRR and aHRs of cancer location in

the bronchiectasis and non-bronchiectasis

cohorts

Table 3 summarizes the cancer locations in the bronchiectasis and non-bronchiectasis cohorts. Patients

with bronchiectasis exhibited a considerably increased risk of lung cancer (aHR = 2.40, 95% CI = 2.22–2.60), oesophageal cancer (aHR = 2.06, 95% CI = 1.61–2.64), and haematologic malignancy (aHR = 2.02, 95% CI = 1.72–2.37) compared with the non-bronchiectasis cohort. In addition, patients with bronchiectasis exhibited a slightly greater risk of head and neck cancer (aHR = 1.54, 95% CI = 1.32–1.80), stomach cancer (aHR = 1.31, 95% CI = 1.13–1.52), colon cancer (aHR = 1.12, 95% CI = 1.01–1.23), hepatoma (aHR = 1.42, 95% CI = 1.29– 1.57), bladder cancer (aHR = 1.43, 95% CI = 1.19– 1.72), thyroid cancer (aHR = 1.54, 95% CI = 1.06–2.24) and other cancers (aHR = 1.32, 95% CI = 1.10–1.59) than did the non-bronchiectasis cohort.

Cumulative incidence of cancer in the

bronchiectasis and comparison cohorts

Figure 1 illustrates the substantially increased cumulative cancer incidence in the bronchiectasis cohort compared with that in the comparison cohort.

Discussion

This is the first study to compare the risks of cancer between patients with bronchiectasis and patients without bronchiectasis in a large population-based cohort in an Asian population during a 14-year follow-up period. The cancer incidence was higher in

the patients with bronchiectasis than in the comparison cohort (17.0 vs. 12.2 per 1000 person-years). The

patients with bronchiectasis exhibited a higher prevalence of comorbidities, namely hypertension, hyperlipidaemia, diabetes mellitus and COPD, than did

the patients in the non-bronchiectasis cohort. However, the patients with bronchiectasis had a 1.46-fold

overall risk of cancer compared with the patients in the non-bronchiectasis cohort after we adjusted for age, sex and comorbidities.

The cause of cancer development among patients with bronchiectasis remains unclear. Bronchiectasis is a chronic respiratory disease characterized by irreversible airway dilatation with inflammation, chronic

bacterial infection, and destruction of bronchial walls (17). Chronic inflammation may be an aberrantly prolonged protective response to a loss of tissue homeostasis (18). Moreover, inflammation can

enable most of the core cellular and molecular capabilities that are required for tumourigenesis (19).

Inflammation may be causally related to cancer

development through processes that involve genotoxicity, aberrant tissue repair, proliferative

responses, invasion and metastases (20). Previous studies have demonstrated that infection causes inflammation-induced tumourigenesis (7,21,22). Approximately 20% of all malignancies worldwide are associated with microbial infection (20,23).

Recent studies have indicated that commensal microbial elements are associated with inflammation and

tumour development (24–28).

In this study, the overall risk of developing cancer was higher in sex-specific, age-specific, and comorbidity-specific bronchiectasis cohorts than in the

comparison cohort. The substantial increase in cancer risk in the bronchiectasis cohort included

increased risks of lung cancer (aHR = 2.40) and oesophageal cancer and hematologic malignancies (aHR = 2.06 and 2.02, respectively). The development of lung cancer may be attributed to chronic

inflammation and recurrent infection of the bronchial trees. Patients infected with M. tuberculosis have an increased risk of lung cancer (29). Pulmonary scarring, which results from various infections, injuries and lung diseases, has been correlated with

developing carcinoma (30). Localized inflammation of the lung related with scarring may induce the development of lung cancer (31). However, systemic

and bronchial inflammation may be present in bronchiectasis patients and may be related to an increased

risk of hematologic malignancies (13). Systemic lupus erythematosus, a chronic systemic inflammatory disease, has been associated with hematologic

malignancies (32). The clinical presentation of bronchiectasis may coexist with gastroesophageal reflux,

which is associated with oesophageal cancer (33). Chest CT images are the routine examinations to diagnose and follow-up bronchiectasis. In addition, lung and oesophageal cancers are the most subsequent cancers (aHR = 2.40 and aHR = 2.06, respectively) of the bronchiectasis cohort in this study.

Therefore, the higher curve’s slope of the cumulative cancer incidence (Figure 1) in the bronchiectasis cohort during the earliest weeks or months partially might be due to increased CT imaging examinations in the bronchiectasis cohort resulted in increasing the detection of lung and oesophageal cancers. The prevalence of bronchiectasis increased with patient age; this result was consistent with the results of previous studies (34,35). Although the cancer incidence increased with age in both cohorts, younger

patients with bronchiectasis had the greatest risk of cancer compared with those in the comparison

cohort. These findings suggest that young adults with bronchiectasis should receive cancer screenings during follow-up.

The strength of this study is that it was a longitudinal nationwide population-based cohort study

investigating the cancer risk in patients with bronchiectasis. The NHI program is universal and mandatory

in Taiwan, and NHI beneficiaries are assigned personal identification numbers that enable them to be traced throughout the follow-up period. Moreover, cancer diagnosis of the patients in our study

was verified based on their inclusion in the RCIPD, indicating that the data were highly reliable. However, certain limitations must be considered when interpreting the findings. First, the NHIRD does not contain detailed personal information, such as data on smoking habits, occupation, lifestyle choices, and family histories of malignancy; this

information might have constituted a potential confounding factor in this study. COPD is a well-established

disease correlated with smoking (36). We adjusted for COPD to minimize the influence of smoking. Second, we included bronchiectasis patients who were hospitalized and possibly presented with more severe symptoms and inflammation. This inclusion may have caused the risk of developing lung cancer to be overestimated. Third, a surveillance bias may have partially contributed to the increased frequency of cancer in the bronchiectasis patients. Lastly, we were unable to retrieve relevant clinical information, such as imaging examination results,

pulmonary function test results and pathologic information, because the identification numbers of the

patients were scrambled.

In summary, this nationwide study of 53,755

bronchiectasis patients with 215,020 person-years of follow-up indicated that bronchiectasis patients had a 1.46-fold increased overall risk of cancer compared with the general population. The findings suggest that clinicians should be aware of bronchiectasis patients’ risk of developing certain cancers. However, because the findings were obtained from an observational study, additional studies should be conducted