Should Health-Care Systems Pay for Replacement Therapy in Patients With Alpha1-Antitrypsin Deficiency? A Critical Review and Cost-Effectiveness Analysis

Stephan A. Alkins, MD and Patrick O'Malley, MD, MPH From the Pulmonary and Critical Care Medicine Service (Dr. Alkins) and the Department of Internal Medicine (Dr. O'Malley), Walter Reed Army Medical Center, Washington, DC.


CHEST. 2000;117(3) 

In This Article


Given the absence of randomized, placebo-controlled trial data with alpha1-AT replacement therapy, and the ethical dilemma of withholding a potentially beneficial treatment in a placebo-controlled study,[21] decisions regarding the use of alpha1-AT augmentation must be based on the best observational data available. Assuming that the mortality rate reduction associated with alpha1-AT replacement therapy in the NIH Registry[10] is valid, this cost analysis suggests that alpha1-AT replacement is cost-effective for severely deficient patients with severe COPD.

Hay and Robin[22] explored the cost-effectiveness of replacement therapy with human alpha1-AT before survival data were available and reached similar conclusions. They explored the cost-effectiveness of replacement therapy under a wide variety of assumptions. LE and survival data were adapted from the reported natural history of alpha1-AT-deficient subjects and from United States vital statistics. Survival was varied depending on age, sex, and smoking history. In their analysis, the efficacy of therapy was defined as the change in survival, and costs were based on medical care cost estimates in individuals with COPD. They concluded that if alpha1-AT replacement therapy had an efficacy of 70%, then the cost per year of life saved would be between $28,00 and $72,000 (depending on patient age, sex, and smoking status). If this therapy were only 30% effective, then the cost per year of life saved would vary between $50,000 and $128,000. The survival advantage in always treated alpha1-AT-deficient subjects with an initial FEV1 <= 50% in the NIH Registry[10] equates to a 55% efficacy (reduction in 5-year mortality rate from 33 to 15%). Our sensitivity analysis yielded results similar to those of Hay and Robin, with the incremental cost per year of life saved varying between $10,747 and $53,735, with an effect size of 55% when yearly alpha1-AT replacement costs vary from $40,000 to $200,000 (Table 1) . This analysis did not address the discounting of costs or quality-of-life changes associated with receiving lifelong IV replacement therapy. While these are limitations, we used a 7% discount rate for the years of life saved, and the findings were not sensitive to this (ie, the treatment was still cost-effective) (Table 2). It is unlikely that these limitations will affect the overall conclusion that replacement therapy is cost-effective as defined by the usual standards.

Although outcome studies such as the NIH Registry data are crucial in justifying any therapeutic modality, the basic rationale for the use of alpha1-AT augmentation is robust. First, severe alpha1-AT deficiency causes emphysema. Second, replacement therapy with alpha1-AT is known to increase the levels of alpha1-AT in serum and epithelial lining fluid.[4,23] This increase in protease inhibitor concentration has been presumed to interfere with the progression of emphysema by protecting against neutrophil elastase.

Third, observational cohort studies have demonstrated a slowing of FEV1 decline in those individuals with severe alpha1-AT deficiency[8,9,10] and improvement in mortality rate[10] when treatment is given. These observational studies reported a statistically significant reduction in the annual FEV1 decline in treated subjects whose initial FEV1 levels indicated moderate to severe airway obstruction, compared to that in untreated subjects. Seersholm et al[8] reported that treated patients with initial FEV1 levels that are 31 to 65% of predicted had declines of 62 mL/yr compared to 83 mL/yr in the untreated group (p = 0.04). The NIH Registry data similarly described a slower annual decline in FEV1 in treated subjects whose initial FEV1 was 35 to 49% of predicted of 66 mL/yr vs 93 mL/yr in untreated subjects (p = 0.003).[10]

Survival in patients with alpha1-AT deficiency has been shown to vary depending on the initial FEV1 and cigarette use. The median overall survival time is 54 years. The median survival time is 49 years for index patients (those who present with alpha1-AT deficiency), 69 years for nonindex patients (those identified by screening), 52 years for smokers, and 67 years for nonsmokers.[24] Data in the NIH Registry demonstrated a strong association between survival and alpha1-AT treatment, even after controlling for potentially important confounding variables. These data showed that survival improves when partial or continuous alpha1-AT replacement therapy is administered to patients with severe alpha1-AT deficiency and severe COPD as detected on the initial FEV1 measurement. Continuously treated patients with initial FEV1 levels <= 50% of predicted had a 5-year mortality rate of 15% vs 33% for untreated subjects (p < 0.001).[10]

The NIH Registry data should be strongly considered given its large study population and good follow-up (84% at 5 years). In addition, the potential bias in this study was conservative. The healthiest subjects (defined by higher initial FEV1 levels, fewer smokers, and greater likelihood of disease being discovered by screening rather than with symptomatic presentations) were more likely to have augmentation therapy withheld, while the treated group, which experienced improved survival, had more severe disease.

While there are no results from randomized, controlled trials that are available to answer the question of whether alpha1-AT replacement therapy impacts on FEV1 decline or mortality (such a trial may be completed in the future),[25] some have suggested that a randomized, placebo-controlled trial would be unethical.[21] The controversy regarding the interpretation of the available studies of alpha1-AT replacement is demonstrated by the differing recommendations by the Canadian and American Thoracic Societies. The Canadian Thoracic Society does not recommend utilizing replacement therapy until a multicenter, placebo-controlled trial demonstrates efficacy.[15] The guidelines of the American Thoracic Society recommend augmentation therapy only for selected individuals (those > 18 years old with PiZZ, PiZ null, or Pi null null phenotypes with alpha1-AT levels < 11 µM/L and abnormal lung function).[26]

Our incremental cost estimates of alpha1-AT replacement, which were based on an ad hoc cost-effectiveness threshold, compare favorably with several other well-accepted therapies. The use of simvastatin for primary prevention of coronary artery disease, for example, has been estimated to cost $195,000 per year of life saved,[27] and mammography screening for breast cancer every 18 months in women aged 40 to 49 years has been estimated to cost $105,000 per year of life saved.[28] Compared to other commonly accepted practices and therapies, alpha1-AT replacement therapy is equivalent to or less expensive than many.

In conclusion, human alpha1-AT replacement is safe and appears to retard the accelerated decline of FEV1 and overall mortality rate in a distinct subset of individuals (PiZZ homozygous and severely alpha1-antitrypsin deficient, with an initial FEV1 <= 50% of predicted). Lifelong augmentation therapy with alpha1-AT in these select individuals is cost-effective by currently accepted standards. Results from a randomized, controlled trial would be needed to fully define whether these observations are accurate or whether they apply to other groups. Such a trial may be unethical or too costly to justify based on the already available data.

Tables 1 and 2 show the results of varied costs and treatment effects. The 5-year mortality rate in untreated alpha1-AT-deficient patients is 33%. A 5% efficacy would result in a 5-year mortality rate of 31.4%, while a 70% efficacy would result in a 5-year mortality rate of 9.9% (see text for the manner of calculation for LE, mortality, and years of life saved).

Example 1
Assuming an annual alpha1-AT replacement therapy cost of $52,000 and a 5-year mortality rate reduction of 55%, alpha1-AT replacement therapy with a 55% efficacy will result in a 5-year mortality rate of 14.9%. This translates into an LEtreated value of 31.1 years; a 33% 5-year mortality rate in untreated subjects translates into an LEuntreated value of 12.49 years. So, LEtreated - LEuntreated = 31.1 - 12.49 = 18.61 years saved. Discounting years of life saved by 7% yields 17.31 years. Cost-effectiveness = cost/years of life saved. An annual cost of $52,000 equates with a 5-year cost of $260,000. The cost-effectiveness in this example (without the discounted number of years of life saved) is calculated in the following manner: $260,000/18.61 years = $13,971 per year of life saved. When the years of life saved are discounted by 7%, cost-effectiveness is calculated as $260,000/17.31 years = $15,020 per year of life saved.

Example 2
Assume that the annual alpha1-AT replacement therapy costs $148,880, that the 5-year mortality rate reduction is 55%, that the years of life saved is discounted by 7% (17.31 years, as shown in Example 1), and that the annual cost of $148,880 equates with a 5-year cost of $744,000. Then, the cost-effectiveness in this example (without discounted number of years of life saved) is calculated in the following manner: $744,000/18.61 years = $40,000 per year of life saved. When the years of life saved are discounted by 7%, cost-effectiveness is calculated as $744,000/17.31 years = $43,004 per year of life saved.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.