Alpha 1-Antitrypsin Deficiency Carriers, Serum Alpha 1-Antitrypsin Concentration, and Non-Small-Cell Lung Cancer Survival

Abstract

Introduction

Although the association between alpha 1-antitrypsin deficiency (α₁ATD) carriers and lung cancer risk has been found, the effects of α₁ATD carriers and serum alpha 1-antitrypsin (α₁AT) concentration on non-small-cell lung cancer (NSCLC) survival remained unclear.

Methods

Patients were selected from the Epidemiology and Genetics of Lung Cancer study at the Mayo Clinic with the criteria of 1) primary NSCLC diagnosis, and 2) available α₁ATD carrier status tested by isoelectric focusing serum α₁AT concentration by immunonephelometry. The effects of carrier status and serum α₁AT concentration on survival were evaluated by Cox proportional hazards models with (1) a landmark approach where overall survival was defined from the time of blood draw to death from any cause, and (2) included only patients with blood draw time prior to initial treatment.

Results

1,321 patients were included in this study, with 179 α₁ATD carriers and 1,142 non-carriers. No differences in overall survival by α₁ATD carrier status were found [adjusted hazard ratio (AHR): 0.98; 95% CI: 0.82-1.18]. However, serum α₁AT concentration was significantly associated with survival among all patients in the landmark model [AHR per 50 mg/dL increments: 1.15; 95% CI: 1.10-1.20] and among patients whose blood were drawn for serum α₁AT level assessment before any treatment [AHR per 50 mg/dL increments: 1.44; 95% CI: 1.21-1.71].

Conclusions

Being an α₁ATD carrier had no significant effect on NSCLC survival. The increased serum α₁AT concentration was a poor prognosis marker for NSCLC, regardless of carrier status.

INTRODUCTION

Lung cancer has been the leading cause of cancer death for both men and women in the United States and many regions of the world for approximately 40 years.¹ Substantial efforts have been made to identify prognostic factors that can be used for better disease management and improvement of survival. To date, only a few prognostic factors such as TNM stage have been established for determining patient management and predicting clinical outcome; however, they still cannot predict individuals’ prognosis accurately within the same stage. Clinicians and researchers have been prompted by this issue to contemplate whether the aggressive nature of lung cancer is genetically predetermined and whether the differences in gene expressions could be identified as additional useful clinical outcome predictors.

Alpha 1-antitrypsin deficiency (α₁ATD) is one of the most common genetic disorders affecting the Caucasian population, especially European descendants. It is an autosomal codominant disorder and characterized by a low serum concentration of alpha 1-antitrypsin (α₁AT).² α₁AT is the main blood-borne serine protease inhibitor of a broad range of proteases, the primary function of which is the inhibition of neutrophil elastase.^3-7 Our previous studies have shown that being an α₁ATD carrier (i.e., heterozygous individuals who carry one deficient allele of the SERPINA1 gene) is an independent risk factor for lung cancer.⁸ The relationship between α₁ATD carrier status and lung cancer prognosis is largely unknown. Several studies have reported that patients with α₁AT expression in their tumor cells had worse prognosis than those without α₁AT expression, suggesting that α₁AT production in tumor cells may relate to more aggressive behavior in some cancers, including lung cancer.⁹ Elevated serum α₁AT concentration has been found in lung cancer patients compared to people without lung cancer.^10,11 However, the association of serum α₁AT concentration with lung cancer prognosis has not been studied. The goals of this study were to investigate whether α₁ATD carrier status and serum α₁AT concentration could affect survival in non-small-cell lung cancer (NSCLC) patients.

MATERIALS AND METHODS

Study population and design

This is a follow up study of our initial report on “Alpha1-antitrypsin deficiency carriers, tobacco smoke, chronic obstructive pulmonary disease, and lung cancer risk” in 2008.⁸ Since 1997, patients with histologically confirmed primary NSCLC have been recruited prospectively into the Epidemiology and Genetics of Lung Cancer study at Mayo Clinic. All study subjects were enrolled following a procedure approved by the Mayo Clinic Institutional Review Board.^{12, 13} Informed consent was obtained from each subject and blood samples for genotyping and serum measurement were collected after recruitment into study. Serum/plasma samples were separated and stored at −80°C. We limited our study samples to 1,321 patients whose α₁ATD carrier status and serum α₁AT concentration have been tested in our earlier study,⁸ by isoelectric focusing¹⁴ and immunonephelometry,¹⁵ respectively. Of the 1,321 patients, 179 were α₁ATD carriers and 1,142 were non-carriers.

Patient follow up and data collection

After initial enrollment for study, all patients were actively followed with subsequent six month and annual follow-up by mailed questionnaires. Annual verification of patients’ vital status was accomplished through the use of Mayo Clinic’s electronic medical notes and registration database, next-of-kin reports, death certificates, and obituary documents filed in patient medical records, as well as the Mayo Clinic Tumor Registry and Social Security Death Index website. Full medical record abstraction was conducted to obtain the following variables for the 1,321 patients: age, gender, race, smoking history including smoking status and pack-years, histologic cell type, TNM stage¹⁶, treatment modality, blood draw time for serum α₁AT measurement, and other medical conditions.

Statistical methods

Patient characteristics were compared between α₁ATD carriers and non-carriers using a Pearson’s χ2 test for categorical variables, and a Student’s t test was used for continuous variables. Overall survival (OS) was defined as the time from NSCLC diagnosis to death. Patients who were alive at the last contact were censored. Univariate and multivariate Cox proportional hazards (PH) models were used to evaluate the prognostic impact on survival of all factors of interest. Factors were included in multivariate models if the P value was < 0.05 in the univariate analysis. Since blood draw times for α₁AT serum concentration varied from baseline to months after diagnosis or treatment, two models were applied to investigate the effect of α₁AT serum concentration on survival. First, a Cox PH model with a landmark approach was used with survival time defined as time from the date of blood draw to death. Patients who had a serum α₁AT value were included in this analysis. In order to eliminate the potential confounding effect of treatment on serum α₁AT concentration, a second model was performed only including those who had their blood draw before the initiation of any treatment. In this model, survival time was calculated as OS described above. HRs associated with a 50 mg/dL increase in serum α₁AT are reported for a clinically useful interpretation. All analyses were performed with SAS software, version 9.2 (SAS Institute, Cary, NC).

RESULTS

Patient characteristics and univariate analysis

Data for this analysis was frozen on August 19, 2010. Median follow-up time for the 326 patients who were alive at the time of analysis was 8.1 years with 90% patients had a follow-up time beyond 5 years. One hundred and thirty two (74%) patients in the carriers group and 863 (76%) in the non-carriers group have died at the time of analysis. The median OS was 2.9 years (95% CI: 2.3-4.0) in the carriers group and 2.7 years (95% CI: 2.4-3.0) in the non-carriers group. Among carriers, there were 106 (59%) MS (including subtypes MS, M1S, and M2S), 44 (25%) MZ (including subtypes MZ, M1Z, and M2Z), and 29 (16%) other uncommon α₁ATD carriers (including M[null], MI, SZ, etc.). For further quality control, serum α₁AT values for carriers whose α₁AT concentration was >250 mg/dL (n=10) and non-carriers >500 mg/dL (n=18) were considered outliers (from Mayo Clinic normal value of 100-190mg/dL) and excluded from all analyses.

Demographic, tumor, and treatment characteristics are listed in Table 1 by carrier status. There were no significant differences in age, gender, race, stage, smoking status, pack years, cell type, and initial treatment between α₁ATD carriers and non-carriers. The median serum α₁AT concentration in carriers was 127 mg/dL (range: 56-249) compared to 178.0 mg/dL (range: 88-496) in non-carriers. There were a total of 135 patients (17 carriers and 118 non-carriers) with blood draw before the initiation of any treatment (10% of the total sample).

Table 1. Characteristics of 1,321 non-small-cell lung cancer patients.

Patient Characteristics	α1ATD Carriers (179)   No. (%)	α1ATD Non-carriers (1142)   No. (%)	P value
Age			0.88
Mean (SD)	65.7 (10.2)	65.4 (11.1)
Median (range)	67.0(37.0-86.0)	67.0 (18.0-92.0)
Gender			0.74
Female	78 (43.6)	513 (44.9)
Male	101 (56.4)	629 (55.1)
Race			0.13
Caucasian	162 (90.5)	1067 (93.6)
Non-Caucasian	17 (9.5)	73 (6.4)
Missing	0	2
Stage			0.79
I	75 (41.9)	440 (38.5)
II	18 (10.1)	109 (9.5)
III	47 (26.3)	314 (27.5)
IV	39 (21.8)	279 (24.4)
Smoking status			0.69
Never	27 (15.2)	169 (15.2)
Former	96 (53.9)	633 (56.9)
Current	55 (30.9)	310 (27.9)
Missing	1	30
Pack years			0.77
Never smokers	27 (15.1)	168 (14.7)
≤40	65 (36.3)	405 (35.5)
>40 and <60	34 (19.0)	255 (22.4)
≥60	53 (29.6)	312 (27.4)
Missing	0	2
Histologic cell type			0.36
Adenocarcinoma	99 (55.3)	682 (59.7)
Squamous	57 (31.8)	305 (26.7)
Others†	23 (12.8)	155 (13.6)
Initial treatment			0.33
Surgery	93 (54.7)	650 (60.0)
Radiation or chemotherapy, or both	60 (35.3)	322 (30.7)
Other or none	17 (10.0)	111 (10.3)
Missing	9	59
Serum α1AT concentration, mg/dL			<0.0001
Mean (SD)	138.6 (42.09)	199.3 (70.99)
Median (range)	127.0 (56.0-249.0)	178.0 (88.0-496.0)
Missing or outliers	10	8

Abbreviations: α₁ATD, alpha 1-antitrypsin deficiency; α₁AT, alpha 1-antitrypsin; α₁AT, alpha-1 antitrypsin; SD, standard deviation.

^†Includes large cell, not specified non-small-cell lung cancer, adenosquamous carcinoma, sarcomatoid, and other mixed cell types.

In univariate analysis, gender, stage, and cell type (P<0.05) were all significantly associated with OS. There was a significant association of serum α₁AT with OS. Carrier status, age, race, smoking status and pack years were not significantly associated with OS (Table 2).

Table 2. Univariate analysis.

Patient characteristics	Unadjusted HR (95% CI)	P value
Age (10 years increase)	1.02 (0.96, 1.09)	0.47
Gender
Female	Reference
Male	1.19 (1.05, 1.35)	0.01
Race
Non-Caucasian	Reference
Caucasian	1.27 (0.98, 1.65)	0.07
Stage
I	Reference
II	1.94 (1.55, 2.44)	<.0001
III	3.15 (2.67, 3.71)	<.0001
IV	5.04 (4.25, 5.96)	<.0001
Smoking status
Never	Reference
Former	0.98 (0.82, 1.18)	0.85
Current	0.90 (0.74, 1.10)	0.32
Pack years
Never smokers	Reference
≤40	0.91 (0.75, 1.10)	0.33
>40 and <60	1.01 (0.82, 1.24)	0.93
≥60	1.03 (0.84, 1.26)	0.77
Histologic cell type
Other cell type	Reference
Adenocarcinoma	0.67 (0.56, 0.81)	<0.0001
Squamous	0.77 (0.63, 0.94)	0.01
α1ATD carrier status¶
Non-carriers	Reference
Carriers	0.93 (0.77, 1.12)	0.43
Serum level of α1AT (50 unit increase)
In all patients†	1.13 (1.09, 1.18)	<.0001
In patients whose blood were drawn  before initial treatment‡	1.39 (1.19, 1.63)	<.0001

Abbreviations: α₁ATD, alpha 1-antitrypsin deficiency; α₁AT, alpha 1-antitrypsin; HR, hazard ratio; CI: confidence interval.

Other cell type includes large cell, not specified non-small-cell lung cancer, adenosquamous carcinoma, sarcomatoid, and other mixed cell types.

^¶N=1321, number of events= 995.

^†Cox Proportional Hazard model with Landmark Approach. Survival time was calculated from the date of the blood draw to death. N= 1303, number of events= 982.

^‡N= 128, number of events= 110.

Multivariate analysis

After adjusting for gender, stage and cell type, no survival differences by carrier status was found in NSCLC (HR: 0.98; 95% CI, 0.82-1.18; P=0.83). A significant association between serum α₁AT concentration and OS was observed in the Cox PH model using the landmark approach (adjusted HR per 50 mg/dL increments: 1.15; 95% CI: 1.10-1.20; P<0.0001) in the presence of gender, stage, cell type, and carrier status. In the subgroup of patients who had their blood draw before treatment, the significant association remained, with an even bigger effect (adjusted HR per 50 mg/dL increments: 1.44; 95% CI: 1.21-1.71; P<0.0001). The results are shown in Table 3.

Table 3. Multivariate analysis.

	Adjusted HR (95% CI)	P value
α1ATD carrier status¶
Non-Carriers	Reference
Carriers	0.98 (0.82, 1.18)	0.83
Serum level of α1AT (50 mg/dL increase)
In all patients‡†	1.15 (1.11, 1.21)	<0.0001
In patients whose blood were drawn  before initial treatment‡	1.43 (1.21, 1.70)	<0.0001

Abbreviations: α₁AT, alpha1-antitrypsin; α₁ATD, alpha 1-antitrypsin deficiency; HR, hazard ratio; CI: confidence interval.

^¶Adjusted for gender, cell type, and stage.

^†Cox Proportional Hazard model with Landmark Approach. Survival time was calculated from the date of the blood draw to death. Adjusted for gender, stage, cell type, and carrier status. N= 1303, number of events=982.

^‡Adjusted for gender, stage, cell type, and carrier status. N= 128, number of events=110.

DISCUSSION

To our best knowledge, this is the first study investigating the roles of α₁ATD carriers and serum α₁AT concentration on NSCLC survival. Although no difference on overall survival by carrier status was found, there were significant survival differences by α₁AT serum concentration in patients with NSCLC, regardless of carrier status. Our study results revealed that potential prognostic role of α₁AT concentration in NSCLC may be clinically useful to both carriers and non-carriers.

At least 10 million Americans and 120 million people worldwide are α₁ATD carriers, who do not normally have severe α₁ATD-related diseases, and most of them are not aware of their carrier status.¹⁷ Although there have been multiple studies investigating the role of α₁ATD carriers and risk of developing cancer, little is known about the potential prognostic impact of α₁ATD carrier status on lung cancer survival. In our analysis of patients with lung cancer, no association between α₁ATD carriers and survival was found. There are several potential explanations. First, the differences between α₁ATD carriers and non-carriers may be too small to affect the overall course of NSCLC in the current study, thus the subtle difference can not be detected even though the sample size of our study was relatively large. Second, the malignancy may have developed mechanisms to act against the effect of carrying an α₁ATD allele. Third, being an α₁ATD carrier may have no impact on NSCLC survival.

α₁AT is the major serine protease inhibitor in human plasma, which is synthesized primarily in the liver and, to a lesser extent, in peripheral blood monocytes, alveolar macrophages, and epithelial cells of the bronchial and gastrointestinal mucosa.^18-20 Researchers have been searching for a correlation between α₁AT and the process of neoplasia, which may help in the diagnosis and the follow up of cancer patients. Various studies have been performed in different types of cancer, such as lung, liver, prostate, pancreases, cervix, and colorectal. Several studies reported an elevation of α₁AT in cancer patients.^{10, 11, 21-25} Additionally, multiple studies have shown some links between α₁AT expression levels and cancer prognosis, but the conclusions were inconsistent.^{24, 26-28} What we currently know about α₁AT led us to believe that this protease inhibitor substance may play a role in NSCLC survival. A higher α₁AT concentration was significantly associated with worse survival in patients with NSCLC in our study. There are several possible mechanisms to explain this finding. First, α₁AT may have a function of modulating host-immunodefence mechanisms in favor of tumor cells, which may suppress the blastogenic or cytotoxic reactions of lymphocytes by inhibiting T cell-mediated cytotoxicity, antibody-dependent cell-mediated cytotoxicity, and natural killer-cell activity.⁹ Second, the participation of α₁AT in tumoral invasiveness seems to be related to its degradation by matrix metalloproteinase, resulting in the production of a COOH-terminal fragment that increases tumor growth in vivo.²⁹ Our findings suggest that α₁AT serum concentration evaluation may be clinically useful in recognizing patients who are more likely to develop a worse outcome in NSCLC. Specifically, a higher serum α₁AT concentration may indicate a tendency for a worse outcome.

Our study has several significant strengths. First, the sample size is reasonably large. Second, the disease phenotype (tumor characteristics) was properly defined. Detailed epidemiological and clinical information were carefully and completely collected, with thorough quality control procedures. Third, the long follow-up time increased the statistical power to detect the effect on survival. However, our study does have limitations. First, although we tried our best to get the patients’ blood samples as soon as possible, still the majority of blood samples could not be drawn before treatment. Therefore when the association between serum α₁AT concentration and NSCLC survival was evaluated, the date of blood draw was used as baseline instead of NSCLC diagnosis to measure survival. We further examined a subgroup of patients with blood draw before treatment, where NSCLC diagnosis date was used at starting point to assess survival. The effect of serum α₁AT on survival became even bigger in these patients (adjusted HR per 50 mg/dL increments: 1.44; 95% CI: 1.21-1.71). In this way, the association between serum α₁AT concentration and NSCLC survival was further demonstrated with a higher α₁AT levels corresponding to worse survival outcome. Second, no data was available on some other potential prognostic factors for survival in the database (e.g., ECOG Performance Status and comorbidity information). Finally, there may be multiple other genes and modulators involved in the SERPINA1 gene mutation, expression, and serum α₁AT concentration, which should be incorporated into the future studies, as well.

In conclusion, we did not find a significant association between α₁ATD carriers and overall survival in NSCLC patients. However, our results indicate that the α₁AT serum concentration may be a biological marker of potential prognostic significance in NSCLC, regardless of carrier status. Our findings will add constructively to the existing body of knowledge about the promising serum biomarkers on NSCLC prognosis. Future studies are warranted to confirm these findings and their clinical application values.