Risk Factors and Clinical Characteristics of Lung Cancer in Idiopathic Pulmonary Fibrosis

A Retrospective Cohort Study

Hongseok Yoo; Byeong-Ho Jeong; Myung Jin Chung; Kyung Soo Lee; O. Jung Kwon; Man Pyo Chung


BMC Pulm Med. 2019;19(149) 

In This Article


Although several studies demonstrated increased prevalence of lung cancer in IPF, only a few studies thoroughly evaluated the cumulative incidence of lung cancer. Ozawa et al. reported 1-, 5-, and 10-year cumulative incidences of lung cancer of 3.3, 15.4, and 54.7% after retrospectively evaluating 103 patients with IPF at their institute.[21] A study by Tomassetti et al. found that lung cancer developed in 23 (30%) of 181 patients with IPF.[12] When calculated only for patients who developed lung cancer, 1- and 3-year cumulative incidences were 41 and 82%, respectively, suggesting most lung cancer in IPF develops within 3 years.[12] Most recently, Yoon et al. identified 31 (2.8%) lung cancer out of 1108 patients with IPF by reviewing their interstitial lung disease registry. Although exact cumulative incidences are not reported in the article, they noted that the incidence of lung cancer was increased in the first 2 years of IPF diagnosis.[11] In addition, the incidence was 3.34 times higher in IPF compared to that of age-adjusted general population. Finally, in a recent systematic review, the estimated adjusted incidence rate ratio from 2 large cohort studies was reported to be 6.42 after adjustment for age, smoking, and gender.[22] The high incidence density rate of 32.6/1000 person-years and 10-year cumulative incidence of 31.1% in our study is in line with previous studies that IPF is frequently associated with lung cancer. Of note, the difference in our study lies in exclusion of patients who were diagnosed with IPF and lung cancer simultaneously and the relatively large number of patients. The study by Tomassetti et al. not only included patients who were diagnosed with lung cancer during follow-up of IPF but also patients who were diagnosed with lung cancer and IPF concurrently which constituted 30% of the 23 lung cancer patients. The study by Yoon et al. also included patients who were diagnosed with lung cancer and IPF at the same time which consisted approximately 20% of lung cancer patients. This inclusion could be the reason for the relatively high early cumulative incidence observed in their studies. Thus, our study may provide more insight on cumulative incidence of lung cancer development in patients during follow up of IPF, which increased persistently over time.

Considering the high incidence of lung cancer and its impact on survival of patients with IPF, understanding the predictive factors of lung cancer development is the first step for clinicians to establish surveillance protocols. Older age at IPF diagnosis, male gender, smoking, and emphysema have been proposed as possible risk factors for lung cancer development in previous studies.[12,21,23,24] Nevertheless, the risk factors of lung cancer still remain unclear because of the small number of evaluated patients, resulting in insufficient statistical power; analysis of characteristics at the time of lung cancer diagnosis; and inclusion of patients concomitantly diagnosed with lung cancer and IPF. In our analysis, male gender, current smoking at the time of IPF diagnosis, and rapid annual decline of FVC were determined to be predictive factors of future lung cancer occurrence.

One of the most noteworthy finding in our study was that rapid decline of FVC was independently associated with lung cancer development. To the best of our knowledge, this is the first study to evaluate and demonstrate the relationship between changes in pulmonary function and risk of lung cancer development in patients with IPF. Although the reason for this association is unclear, the suggested mechanism of frequent lung cancer development in IPF may explain this observation. A longstanding hypothesis is that shared pathogenesis of tissue damage and abnormal repair, which are key processes in IPF and lung cancer development, is the reason that patients with IPF are vulnerable to cancer occurrence.[25] The finding that lung cancer arises commonly abutting or within fibrosis may support this hypothesis. Recent advances in molecular techniques also offer genetic- and epigenetic-level evidence that abnormal DNA methylation and histone modification leading to abnormal gene expression or aberrant activation of signaling pathways are shared by both IPF and cancer.[26] Thus, patients with rapid decline in FVC, which reflects the active and progressive status of IPF, may be more prone to lung cancer development, considering the common pathogenesis of IPF and lung cancer. However, further studies are warranted to confirm this hypothesis.

The results of our study on demographic, clinical, histologic, and radiologic characteristics of patients with IPF and lung cancer did not differ from previous studies. Most patients were older men with a current or ex-smoking history.[12,13,21] Median time from diagnosis of IPF to lung cancer was 38 months, similar to 30 months in a study by Tomassetti et al..[12] Median FVC and DLco were slightly lower than in previous studies. This difference may be because we included only patients with a diagnosis of lung cancer during follow-up for IPF, thus excluding patients diagnosed incidentally with pulmonary fibrosis who might have been in relatively early stage IPF.[7,13,21] Previous studies demonstrated that squamous cell carcinoma is the most common histologic type encountered in IPF.[10,21,26] Our data concurs, with 32.6% of squamous cell carcinoma. Radiological findings were in line with previous studies with round or ovoid masses frequently observed abutting or within fibrosis.[27,28] Although both upper and lower lobe predilections have been described,[27,29] lesions in lower lobes were more frequent in our study.

One of the radiologic findings that requires an attention in our study is that the proportion of stage I lung cancer was statistically higher in patients whose interval of chest CT scans was 1 year or less. Despite the frequency of lung cancer and its detrimental impact on prognosis in patients with IPF, currently there is no screening protocol available for these patients.[1] Although the guidelines recommend lung cancer screening with annual low dose chest CT scan in high risk patients,[30,31] the benefit of this recommendation has not been validated in patients with IPF. The result of our study may suggest a possibility that annual or shorter-term CT screening may also be beneficial in early detection of lung cancer in patients with IPF. Nevertheless, the retrospective observational nature of this study and various reasons which may have initiated short-term CT scan follow-up limit the interpretation of the results. In addition, considering the high rate of treatment related complications and possible non-eligibility for curative treatment due to poor lung function in patients with IPF,[13,32] the survival benefit must be scrutinized as well. Further studies are necessary to determine the benefit of the lung cancer surveillance with chest CT scan and optimal duration of follow-up for patients with IPF.

Significant reduction in survival for patients with lung cancer and IPF compared to that of patients with IPF alone was portrayed in our study. Although Ozawa et al.[21] noted no difference in survival between IPF patients with and without lung cancer, most recent studies consistently reported worse survival in patients with lung cancer.[11,12] This detrimental impact of lung cancer on IPF is known to be largely attributable to progression of lung cancer or treatment-related complications.[12] The high incidence and effects of lung cancer on patients with IPF emphasizes the importance of establishing a surveillance protocol for early diagnosis as well as defining effective and safe treatment modalities for those patients.

Potential limitations should be acknowledged to fully appreciate the results of our study. First, given the observational nature, there is always the possibility that selection bias of confounding might have influenced our findings. In addition, our study was conducted at a single referral institution with an interstitial lung disease clinic, which may limit its generalizability to other settings. Second, we were not able to determine the association between use of antifibrotic agents and risk of lung cancer development. A recent study reported that treatment with pirfenidone was associated with reduced incidence of lung cancer in patients with IPF.[33] However, since our study consisted of patients diagnosed between 1998 and 2005, only a small proportion of the study patients received pirfenidone or nintedanib which limited us from further analysis.