Short Communication Alpha1-Antitrypsin Deficiency Allele Carriers among Lung Cancer Patients

Lung cancer (LC) and chronic obstructive pulmonary lung diseases (COPDs; including emphysema and chronic bronchitis) share a common etiology. Despite the known associations of alpha1-antitrypsin deficiency (a1AD) with COPD and COPD with LC, few studies examined the association of a1AD alleles and LC. We hypothesize that heterozygous individuals who carry a deficient allele of the a1AD gene Pi (protease inhibitor locus) are at an increased risk of developing LC. The Pi locus is highly polymorphic with >70 variants reported. There are at least 10 alleles associated with deficiency in alpha1antitrypsin. Using an exact binomial test, we compared the a1AD carrier rate in 260 newly diagnosed Mayo Clinic LC patients to the reported carrier rate in Caucasians in the United States (7%). a1AD carrier status, determined by isoelectric focusing assay, was examined with respect to the history of cigarette smoking, COPD, and histological types. Thirty-two of the 260 patients (12.3%; 95% confidence interval, 8.6–16.9%) carried an a1AD allele, which was significantly higher than expected (P 5 0.002). Twenty-four of the 32 carriers had allele S, 6 had allele Z, and 2 had allele I. Patients who never smoked cigarettes were three times more likely to carry a deficient allele (20.6%; P 5 0.008), although smokers had a higher carrier rate (11.1%; P 5 0.025) when compared with the 7% rate. Patients with squamous cell or bronchoalveolar carcinoma had a significantly higher carrier rate than expected (15.9% and 23.8%, P < 0.01, respectively). Our preliminary findings suggest that individuals who carry an a1AD allele may have an increased risk for developing LC, specifically squamous cell or bronchoalveolar carcinoma. Introduction An unresolved paradox of LC is that, although a majority of the patients are tobacco users, only a minority of long-term smokers develop LC. In addition to cigarette smoking, another well-known risk factor for LC is COPD, which includes emphysema and chronic bronchitis (1). COPD not only increases LC risk in both smokers and nonsmokers, but also shares common risk factors with LC (1–4). Both diseases are strongly associated with tobacco use (1, 5), and both aggregate in families (4–6). Individuals who are homozygous for the a1AD gene (7) or who are heterozygous for this gene (a1AD carriers) are predisposed to the development of COPD (8–10). However, it is not known whether a1AD individuals and carriers are at increased risk of LC. a1AT, a secretory glycoprotein produced in the liver, is a protease inhibitor and neutralizes the effects of proteases in several organ systems, mainly the lung. It is believed that COPD develops in a1AD individuals as a result of an imbalance between neutrophil elastase (a protease) and a1AT in lung tissue (8, 11, 12). This imbalance could be due to an excess of elastase and/or a lack of functional a1AT (8, 13). Normal plasma a1AT concentration or level is 110–200 mg/dl, and a1AD individuals have a1AT levels ranging from 0–60 mg/dl (7, 14). The a1AT level is marginally normal in those who are heterozygous for the deficient alleles (70–110 mg/dl; Refs. 14 and 15). Tobacco smoke disturbs the balance between protease and protease inhibitor activity in lung tissue by stimulating neutrophils to secrete more elastase (16) and inactivating a1AT (17, 18), thereby leading to elastolytic destruction of lung tissue (19). Because of debilitating consequences (7, 14, 19), individuals with known a1AD usually have minimal or no tobacco smoke exposure (20). a1AD carriers do not normally suffer from severe a1AD-related diseases; however, they may be especially vulnerable to tobacco smoke-related diseases. Whether an individual with elastolytically destroyed lung tissue is more susceptible to respiratory carcinogens has not been properly evaluated. As an initial step to test this hypothesis, we attempt to answer the following question: Are LC patients more likely to carry a1AD alleles than the general population? Materials and Methods Study Subjects. Patients were selected from an ongoing comprehensive LC study at the Mayo Clinic in Rochester, Minnesota. The goals of this study are to examine the roles of genes and environmental exposures in LC etiology, to search for susceptibility and risk-modifying genes, and to identify at-risk individuals. Since March 1997, all patients who were newly diagnosed with primary LC have been identified and evaluated Received 6/30/98; revised 12/7/98; accepted 3/8/99. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Partially funded by Mayo Clinic/Foundation and National Cancer Institute Grant CA77118. 2 To whom requests for reprints should be addressed, at Mayo Clinic and Foundation, 200 First Street, SW, Rochester, MN 55905. Phone: (507) 266-5369; Fax: (507) 284-1516; E-mail: [email protected]. 3 The abbreviations used are: LC, lung cancer; COPD, chronic obstructive pulmonary diseases; a1AT, alpha1-antitrypsin; a1AD, alpha1-antitrypsin deficiency; BAC, bronchoalveolar carcinoma; CI, confidence interval; FH, family history. 461 Vol. 8, 461–465, May 1999 Cancer Epidemiology, Biomarkers & Prevention on June 21, 2017. © 1999 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from for study eligibility. Our eligibility criteria were developed to efficiently achieve our goal of a balanced case group by oversampling cases with characteristics indicative of other risk factors besides smoking: (a) all cases with a positive FH of LC in at least one first-degree relative, and all cases with a positive FH of other cancers in at least two first-degree relatives; (b) all cases who are diagnosed at age 50 or younger; (c) all cases who are lifetime nonsmokers (smoked ,100 cigarettes during lifetime); (d) all cases who have uncommon tumor types; and (e) a 20% sample of those LC patients who do not meet criteria 1–4 (i.e., the sporadic group). Group 5 can serve as an internal comparison for groups 1–4 and can be used to derive a complete patient ascertainment. The sporadic cases were selected in such a way that every fifth sporadic case identified was enrolled to achieve a 20% sample of the patients in this category. The rationale for our sampling strategy is as follows: although a majority of all cases are cigarette smokers, LC presents as a very heterogeneous disease with regard to clinical presentation, histopathological features, age at diagnosis, and FH of cancer. On average, one-half of the patients are so-called “sporadic” cases who seem to be typically smoking-related, develop a common type of LC after 50 years of age, and have no significant FH of lung or other cancer. Common types of LC include adenocarcinoma, squamous cell carcinoma, small cell carcinoma, large cell carcinoma, and unspecified non-small cell LC. New patients were ascertained daily from our computerized pathology reporting system, which identifies approximately 95% of all LC cases seen at our institution. The remaining 5% of the patients were identified directly from the patient-care physicians. Medical records for each patient were reviewed to determine study eligibility, except for a small number of patients who denied research authorization for review of their medical records (,1%). After informed consent, each patient was interviewed by a certified genetic counselor (K. A. W.) for complete FH (a five-generation pedigree). Patients who were classified as nonsmokers were confirmed at the interview, and a detailed passive smoking history was obtained. From each consenting patient, a 28-ml blood sample and environmental exposure information (tobacco consumption, dietary pattern, and occupational exposures) were requested for study. Due to the dynamic nature of an ongoing study, this report presents preliminary results based on cases identified during the first 9 months of our study. Between March and November 1997, 597 eligible patients were identified. Among these, 400 (67.3%) were interviewed and 68 (11.4%) declined. Three hundred eighty-eight of the interviewed patients (97%) agreed to donate blood, 302 blood samples have been received, and a1AT allele types were tested on 260 patients in the study period defined for this early report. Because the a1AD allele frequency has been well documented worldwide and varies greatly from Caucasians to other populations (7, 21–23), only patients of European origin were included in the current report. Among the 42 patients (i.e., the difference between the 302 and 260), for whom we have collected blood samples but were not included in this early report, 3 were non-Caucasian of European origin and 39 did not have a pathological report from our institution. Data Collection. Information abstracted from the patients’ medical records includes pathology, clinical staging, treatment, history of previous diseases, lifestyle (tobacco, alcohol, and coffee use), education, occupation, a brief FH, demographics, and follow-up data. History of COPD was based on explicit diagnosis (24) or abnormal pulmonary function tests (25) that were recorded in the patients’ medical histories. During the patient interview, a five-generation pedigree was constructed for each patient. Data collected for each relative in the pedigree included: vital status and health history concerning malignant and nonmalignant diseases (age at diagnosis and/or cause of death). Information on tobacco use and major occupation was collected for all first-degree relatives. a1-AT Allele Determination and Plasma Concentration Measurement. a1AD is a common autosomal recessive disorder caused by mutations of the protease inhibitor locus Pi, located on chromosome 14q32.1 (7, 21). Pi is highly polymorphic with .70 variants reported (7, 26, 27). Each variant is designated by a letter corresponding to the migration of the a1AT protein in isoelectric focusing assay, the standard clinical diagnostic test for more than 20 years (28–33). M (including subtypes M1, M2, and M3) is the most common, with most normal individuals being homozygous for this allele (designated as MM or PiMM with the subtypes, e.g., M1M1). Of the variants that lead to a deficiency of a1AT, only Z and S alleles are common. Uncommon deficient alleles include I, MMalton, MPittsburgh, null, and other rare alleles (7). Isoelectric focusing assay, performed in the Mayo Clinic Protein and Immunopathology Laboratory (33, 34), was used to type a1AT alleles, and the concentration of plasma a1AT was determined by nephelometry using standard protocol by Beckman Instruments, Inc. (35). Data Analysis. An exact binomial test (36) was used to compare the a1AD carrier rate among LC patients with the expected frequency of 7%. An observed to expected ratio was calculated. Although Pi allele frequencies vary substantially across geographic regions and ethnic groups in Europe, they are fairly homogeneous in the United States white population (22, 23, 37, 38). The Z allele is found in 1–2%, and the S allele in 2–4% of all Caucasians of European descent, but ,1% for both alleles combined in Asians and Africans (7, 8, 23, 39). Previous studies in Minnesota populations reported a prevalence of 1.4% for the Z allele and 2.3% for the S allele (37–39). Assuming Hardy-Weinburg equilibrium (40), the proportion of heterozygous individuals is estimated at 7%. The I allele, which causes moderate a1AD (60–70% of normal level), is very rare with a frequency of ,0.003 in United States whites (38, 41). Results were also stratified by history of COPD and tobacco smoking (coded as binary variables). Smoking status was divided into neverand ever-smoked in this analysis. x statistics were used in comparing carriers to noncarriers for smoking and COPD history, and the Wilcoxon rank sum test was used to compare ages and a1AT levels of the carriers and noncarriers (42). The 95% CIs are exact intervals based on Feller (43).