Smoking is not associated with lung metastasis in pancreatic cancer

Abstract

Smoking is a risk factor for pulmonary metastasis in various malignancies. We investigated this association for pancreatic ductal adenocarcinoma (PDAC). We conducted a retrospective 1:2 case-control study of consecutive patients who underwent PDAC resection (2011–2021). Cases ultimately developed lung metastases and controls did not. Of 744 patients we identified 53 cases and 106 matched controls. Twenty-five (47%) cases and 50 (47%) matched controls had a history of smoking (p=1.0). This indicates that smoking is not associated with increased risk of pulmonary metastasis in resectable PDAC. Further research is needed to elucidate tumor and parenchymal factors influencing metastatic site.

Introduction:

In 1889, Stephen Paget proposed the ‘seed and soil’ hypothesis, postulating that metastasis depends on cross-talk between cancer cells and specific organ microenvironments[1]. Subsequently, numerous studies have attempted to elucidate why each tumor type has a predilection for specific metastatic sites, and how environmental factors alter this predisposition. Studies have identified cigarette smoking as a risk factor for pulmonary metastasis in breast cancer and several gastrointestinal malignancies[2–4]. Literature review did not identify studies that investigated the association between cigarette smoking and likelihood for lung metastasis in pancreatic ductal adenocarcinoma (PDAC). We sought to understand the relationship between cigarette smoking and lung recurrence in patients with PDAC as cigarette smoking is estimated to be a risk factor for approximately one-third of PDAC cases.[5] Additionally, an association between smoking and lung metastasis has prognostic significance. Patients with initial lung metastases live twice as long as those with initial liver or peritoneal metastasis (30.4, 15.0, 14.1 months respectively)[6].

We hypothesized that a history of smoking would be a risk factor for initial pulmonary recurrence in patients with PDAC. We investigated this association using a retrospective cohort of patients who underwent PDAC resection at Columbia University Irving Medical Center (CUIMC) from 2011 to 2022.

Materials and Methods:

We conducted a retrospective review of 744 consecutive patients who underwent resection of stage I-III PDAC from January 2011 through December 2021 at CUIMC. Data collected included age at diagnosis, surgical and pathologic characteristics, site and date of initial recurrence, time to diagnosis or last follow-up. Surgical staging was performed using the American Joint Committee on Cancer (AJCC) 8th edition staging system. Survival analyses were performed using Kaplan-Meier statistics.

For each patient with lung as the site of initial recurrence, we matched two patients who did not develop pulmonary recurrence during the same follow up period (both those without metastasis and those with metastasis to non-lung sites). Matching was initially based on race and sex. Patients were then matched to prospective matches with age at diagnosis closest to the primary lung patient. Data on smoking had not been collected at this point and the investigator identifying matched patients was blinded to the smoking status of the patients. We then reviewed all matched triplets for smoking history, including history of current or prior smoking and of significant environmental exposure to smoke. We performed a chi-squared test to determine association between history of smoking and development of pulmonary metastases.

Based on the number of patients in our study, this study had an 80% power to identify a 25% difference in the rates of smoking between these two patient groups.

The research was approved by the Institutional Review Board (IRB) at CUIMC (IRB AAAS5206).

Results:

We identified 744 consecutive patients with resected PDAC at CUIMC, of whom 53 (7.1%) developed an initial lung recurrence. These patients were matched with 106 patients without initial pulmonary recurrence. The characteristics of these patient groups are described in Table 1. Median age was 68 years and males comprised 57% of patients in both groups. Sites of first recurrence in the control group included liver (25), local (24), peritoneum (9) and brain (1), including patients with multiple-site initial recurrence.

Table 1:

Patient Demographics

		Lung Recurrence N (%)	Control N (%)
Total		53 (100)	106 (100)
Median age at diagnosis (years)		68	68
Sex	Female	23 (43)	46 (43)
	Male	30 (57)	60 (57)
Ethnicity	Hispanic	6 (11)	13 (12)
	Not Hispanic	47 (89)	93 (88)
Race	White	46 (87)	88 (83)
	Black	3 (6)	7 (7)
	Asian	2 (4)	4 (4)
	Other/omitted	2 (4)	7 (7)
Never Smoker	Total	28 (53)	56 (53)
Smoking History	Total	25 (47)	50 (47)
	Former	22 (42)	42 (40)
	Current	1 (2)	8 (8)
	Significant environmental exposure	2 (4)	0 (0)
Neoadjuvant Chemotherapy	Yes	17 (32)	32 (30)
	No	38 (72)	74 (70)
AJCC surgical T stage at resection	T0	0 (0)	2 (2)
	T1	2 (4)	6 (6)
	T2	11 (21)	22 (21)
	T3	38 (72)	75 (71)
	T4	2 (4)	1 (1)
AJCC surgical N stage at resection	N0	7 (13)	34 (32)
	N1-N2	46 (86)	72 (68)

Median follow up time was 2.6 years in the lung-recurrence group and 1.6 years in the control group. Median time to recurrence was 17.2 months in the cases and 11.4 months in the control group. One- and two-year survival rates were 91% and 73% in patients with lung metastases and 84% and 62% in the control group.

Of the 53 patients with pulmonary recurrence, there were 25 (47.2%) patients with a history of smoking, including 1 current smoker. Of the 106 controls, there were 50 (47.2%) patients with a history of smoking, including 8 current smokers. A chi-squared test for association did not find a statistically significant difference in the rates of smoking history between the groups (p = 1.0). In patients with liver metastasis in the control group, 52% had a history of smoking.

Discussion:

Smoking has been identified as a risk factor for pulmonary metastasis in several malignancies. This association was first described in breast cancer, in which case-control studies demonstrated increased rates of both current and former smoking in patients with lung metastases than in those without. A large prospective cohort study of patients with breast cancer found that every 1000 packs of cigarettes consumed over a lifetime increased a woman’s risk of developing lung metastases by about 3–7% (p < 0.001)[2]. A retrospective review of 567 patients with stage I-III colorectal cancer, 39 (6.9%) of whom went on to develop pulmonary metastases, found that current smoking was an independent risk factor for the development of pulmonary metastases[3].

The mechanism for the association with lung metastases is not fully understood. One possible theory posits that smoking-induced inflammation increases pulmonary capillary permeability and the expression of cellular adhesion molecules, allowing cancer cells to enter and colonize lung tissue[7]. In a cohort of 400 patients with breast cancer, including 100 with lung metastases, smoking history and lung metastasis were both strongly associated with increased levels of prostaglandin E₂, supporting a relationship between smoking-induced inflammation and pulmonary metastasis[8]. Additionally, pulmonary macrophage-specific matrix metalloproteinases, which degrades the extracellular matrix, may have increased activity in smokers, leaving the lung parenchyma vulnerable to penetration and colonization by tumor cells[9]. Smoking-induced changes in natural killer cell and platelet function may also increase rates of metastasis regardless of site[7].

The results of our study indicate that smoking is not a risk factor for the development of pulmonary metastases in patients with PDAC. This highlights the complex, multifactorial tumor-specific and parenchyma-specific factors which may lead a particular tumor to metastasize to a specific location. Further research must be conducted to further understand the tumor-environment for PDAC and the driving factors behind location of recurrence.