Evaluating the financial impact of utilizing recombinant porcine factor VIII or recombinant FVIIa for patients with acquired hemophilia A

ABSTRACT

Objective

To evaluate the financial impact of utilizing rpFVIII or rFVIIa during a hospital admission for the diagnosis of acquired hemophilia A (AHA) by reviewing the margin between the cost to the hospital for providing care and the amount the hospital is reimbursed by the Centers for Medicare & Medicaid Services (CMS) in the US.

Methods

Data source was the Medicare Limited Data Set, which contains claims for hospitalizations, charges, and amounts reimbursed by CMS. Study patients were hospitalized with AHA and treated with rpFVIII and/or rFVIIa between 1/1/2015 and 12/31/2019. CMS Fiscal Year 2020 Impact Files, with hospital-level cost-to-charge ratios (CCRs), were used to estimate hospital costs. Sensitivity analyses were conducted to estimate margins at different CCRs.

Results

Hospital margins were, on average, positive with use of either rpFVIII or rFVIIa (rpFVIII: $51,548.89; rFVIIa: $35,943.80). Sensitivity analysis results suggest that the use of rpFVIII is similiar, compared with rFVIIa for a large majority of hospitals.

Conclusions

While there may be higher reimbursement for rpFVIII hospitalizations, this analysis suggests that the use of rpFVIII, compared to rFVIIa, may have no impact on hospital finances for the majority of hospitals, despite rpFVIII’s higher per unit cost.

KEYWORDS:

• Acquired hemophilia
• bypassing agents
• financial impact
• cost-to-charge ratio
• hospital revenue
• hospital charges
• hospital costs
• medicare
• rare disease
• recombinant porcine FVIII
• hospital finances

1. Introduction

Acquired hemophilia A (AHA) is a rare autoimmune disorder that results in sudden, and oftentimes uncontrolled, bleeding in individuals [1]. AHA is caused by the spontaneous inhibition of clotting factor VIII (FVIII) by the production of inhibitors or autoantibodies [1]. Though AHA largely affects the elderly, it can also present in younger patients due to pregnancy, autoimmune diseases, and malignancies [2]. The treatment of spontaneous bleeding in AHA patients, or hemostatic management, can be difficult to manage.

There are currently two types of first-line treatments that are used to control bleeding in patients with AHA: clotting factor FVIII replacement therapy and bypassing agents (BPA) [1,3]. Obizur®, or recombinant porcine factor VIII (rpFVIII), is a factor VIII replacement therapy that was approved for the treatment of AHA by the U.S. Food and Drug Administration (FDA) in 2014 [1,4]. NovoSeven®, or activated recombinant factor VII (rFVIIa), was approved for the treatment of AHA in 1996 and is currently the only BPA that has received FDA approval for AHA treatment in the U.S., though activated prothrombin complex concentrate (aPCC) is another BPA that is sometimes used in an off-label manner to mitigate AHA-related bleeds [1,4,5].

Despite the introduction of rpFVIII as a treatment for AHA, well-known insight from practicing hematologists in the U.S. indicates that rpFVIII is not readily available at most institutions [6]. The limited use of rpFVIII may be due to the perceived impact on hospital finances; it may also reflect the long-standing use of BPAs which has led to provider familiarity and comfort [6]. Hospitals may instead prefer to stock and prioritize use of lower per unit cost BPAs rather than the higher per unit cost rpFVIII. However, the perceived high per unit cost of rpFVIII does not necessarily mean that it confers an increased economic burden to hospitals’ finances. Further, recent findings suggest that treatment with rpFVIII may reduce risk of mortality or readmissions in the hospital setting compared to the use of rFVIIa [7]. For example, Batt et al. found that the probability of death or readmission with treatment within 12-months of a previous treatment was 40% lower in patients treated with rpFVIII than patients treated with rFVIIa alone (p = 0.047, adjusted for age, sex, and Charlson Comorbidity Index) [7]. Given the evidence demonstrating clinical benefit of rpFVIII, it is important to understand how the use of rpFVIII impacts a hospital’s finances in order to provide hospitals with data-based evidence to inform hospitals in optimal formulary decisions.

To better discern a hospital’s margins after use of rpFVIII to treat AHA, it is necessary to understand the costs of providing treatment as well as the costs recouped from reimbursement. Patients with AHA are predominantly treated in the hospital setting due to the acute, severe, and unpredictable nature of AHA-related bleeds. The Centers for Medicare & Medicaid Services (CMS) assign the hospitalization a diagnosis-related group (DRG) which is then used for payment to the hospital. The DRG is based on the chief diagnosis for which the patient was admitted, patient comorbidities and complications. Medicare then pays hospitals an amount based on the DRG. Hospitalizations where patients are more seriously ill, suffer from more profound comorbidities and complications, are reimbursed at higher rates. Nonetheless, the amount reimbursed by Medicare is typically substantially lower than the charged amount submitted by the hospital. While the charged amount may not reflect the actual costs to the hospital of providing the hospitalization, assessing the costs to the hospital and the Medicare payments for a hospitalization with an rpFVIII administration is needed to understand rpFVIII’s impact on hospital margins. Medicare provides hospital-specific cost-to-charge ratios (CCR) that allow the researcher to understand the relationship between the costs a hospital incurs compared to what the hospital charges. For example, a CCR of 0.22 means that the hospital cost is 22% of what is being charged. Using CCRs, both the charged amount and hospital cost can be compared directly to reimbursed amount to understand a hospital’s margins and the visit’s overall financial impact to the hospital.

We conducted this analysis to assess the impact of the use of rpFVIII in the treatment of patients hospitalized for AHA on a hospital’s finances. Though the per unit cost of rpFVIII is perceived to be higher than that of other first-line treatments such as rFVIIa, the financial implications to hospitals from its use cannot be determined through acquisition costs alone. In this study, we investigate the financial impact of utilizing rpFVIII during a hospital admission and/or rFVIIa during a hospital admission by reviewing the margin between the cost to the hospital for providing care and the amount they are being reimbursed for those services.

2. Methods and study design

2.1. Patient population and data sources

This analysis used the Medicare Limited Data Set (Medicare LDS) (2015–2019) – a deidentified, longitudinal US Medicare database made publicly available to researchers by the Centers for Medicare & Medicaid Services (CMS) – to identify patients with AHA who were treated with either rpFVIII and/or rFVIIa in the US. The dataset contains enrollment,demographic information, and healthcares servies for all individuals in the US with fee-for-service insurance coverage through the publicly funded Medicare program. Covered individuals are eligible, through criteria of age and/or disability as the program provides coverage to individuals 65+ and to individuals under age 65 who have a demonstrated disability; however, the large majority of covered individuals are age 65 and above. The dataset also includes all adjudicated (processed and paid) claims from healthcare providers, for medical services provided to Medicare beneficiaries, and reimbursement for hospitalizations is made relative to the patients’ level of illness and reason for hospitalization. In the recently published study that aligns with this research, the identified patients were then followed for 12-months post-treatment to observe mortality and readmission rates. Patient and admission details can be found in Tables 1 and 2 and detailed patient selection methodology and real-world clinical outcomes are described in Batt et al. and the Supplemental Appendix [7].

Table 1. Patient level baseline demographics and other patient characteristics [7].

Patient Characteristics	rpFVIII (n = 25)		rFVIIa (n = 114)		Total Population (n = 166)
	Mean	SD	Mean	SD	Mean	SD
Charlson Comorbidity Index score	2.32	2.29	1.74	1.97	1.83	2.02
Age	74.96	6.58	74.43	10.60	74.5	10.08
	Observed Difference* between rpFVIII (n = 25) and rFVIIa (n = 141)				N	%
Male	+0.88%				105	63.25%
Race
White	−13.3%				125	75.3%
Black	+13.1%				28	16.9%
Other	+2.58%				13	7.8%
Region
Midwest	+10.0%				39	23.5%
Northeast	+2.0%				37	22.3%
South	+10.0%				59	35.5%
West	−22%				31	18.7%
CCI Component Prevalence at Baseline
Cerebrovascular Disease	−1.93%				16	9.64%
Congestive Heart Failure	−1.73%				29	17.47%
Chronic Pulmonary Disease	−5.99%				35	21.08%
Dementia	−1.67%				-	-
Diabetes	+7.69%				29	17.47%
Hemiplegia or Paraplegia	+1.87%				-	-
HIV/AIDS	+0.71%				-	-
Myocardial Infarction	+6.78%				17	10.24%
Metastatic Solid Tumor	+5.87%				-	-
Peptic Ulcer Disease	−1.67%				-	-
Peripheral Vascular Disease	−10.89%				22	13.25%
Renal Disease	−4.82%				40	24.10%
Rheumatologic Disease	+7.23%				23	13.86%
Mild Liver Disease	+11.74%				-	-
Moderate/Severe Liver Disease	−1.42%				-	-
Other Comorbidities
Autoimmune disorders^†	+12.77%				148	89.16%
Medications^‡
Treated with factor VIII	−11.73%				29	17.47%
Treated with aPCC	−12.82%				38	22.89%

*Calculated as the percent of the rpFVIII cohort with the characteristic of interest minus the percent of the rFVIIa cohort with the characteristic of interest. Positive values indicate higher prevalence in the rpFVIII cohort.

^†Autoimmune lymphoproliferative syndrome, hemophagocytic lymphohistiocytosis, Langerhans cell histiocytosis, acute lymphoblastic leukemia, chronic lymphocytic leukemia, multiple myeloma, Wiskott–Aldrich syndrome, non-Hodgkin’s lymphoma, rheumatoid arthritis, Sicca (Sjögren) syndrome, connective tissue diseases, systemic lupus erythematosus, thyroiditis, antiphospholipid syndrome, diseases of the blood and blood-forming organs, and certain other disorders involving the immune mechanism.

^‡Patients were assessed for the presence of any claims indicating treatment with medications of interest and/or presence of relevant comorbidities at any time from the pre-index period to the end of their follow-up. “-” Values with <11 patients masked to comply with data privacy best practices. Charlson Comorbidity Index = CCI; aPCC = Activated prothrombin complex concentrate; rFVIIa = activated recombinant Factor VII; rpFVIII = recombinant porcine FVIII.

Table 2. Admission level findings [7].

Outcome	rpFVIII		rFVIIa
Total Number of Admissions	40		180
Mean Length of Stay (LOS) (SD)	17.2 (14.4)		15.2 (16.6)
	rpFVIII vs. rFVIIa Admissions
	Unadjusted Results		Adjusted* Results
	Probability	p-value	Probability	p-value
Death during most recent admission	−12.5%	0.075	−13.0%	0.059
Death within 90 days post-discharge of most recent admission	−20.2%	0.025	−21.4%	0.014

Note: rFVIIa admissions served as the reference group. For example, a − 12.5 probability of death during the most recent index admission means that patients treated with rpFVIII had a 12.5% lower probability of dying during the hospital admission than patients treated with rFVIIa.

*Calculated using multivariate regression, adjusted results were controlled for age, gender, and Charloson Comorbidity Index score.

rFVIIa = activated recombinant Factor VII; rpFVIII = recombinant porcine FVIII.

    

To identify hospital-level CCRs, the Fiscal Year (FY) 2020 Impact Files from CMS were utilized. Additional information on how these files were used can be found below. Impact files are typically created in the summer preceding the Federal Fiscal year and rely on the best available data at that time. They can be used by hospitals to estimate the payment impacts of various policy changes to the inpatient prospective payment system (IPPS) that occur as noted in the Federal Register [8]. The FY 2020 Impact File uses data from FY 2017 Healthcare Cost Report Information System (HCRIS); thus, there is a data lag of three to 4 years in reported CCRs.

2.2. Study measures

This study reviews AHA hospital admissions treated with rpFVIII or rFVIIa.

2.2.1. Primary objective

The primary objective of this study was to determine the financial impact of inpatient admissions treated with rpFVIII or rFVIIa by calculating the difference between hospital cost and reimbursed amounts for each of those admissions. Reimbursed amounts represent the actual amount that Medicare pays to the hospital; reimbursed amounts do not include patient cost-sharing responsibilities such as copayments and deductibles.

2.3. Financial analyses

To better understand the financial impact of utilizing rpFVIII or rFVIIa, we compared hospital costs for admissions with rpFVIII or rFVIIa to the amounts CMS reimbursed hospitals for these admissions. To do so, we identified hospital charges and reimbursed amounts recorded in Medicare claims. Because reimbursed and charged amounts were from different years, we adjusted these amounts to 2020 dollars using the medical care component of the Consumer Price Index (CPI). We then used CMS’s FY 2020 Impact file, which contains the cost-to-charge ratio (CCR) for each hospital, to determine hospital costs.

The CCR is the ratio of each hospital’s costs to the charges each hospital records on claims it submits to CMS for reimbursement. The costs in the CCR reflect institution-specific cost information that hospitals that are eligible for reimbursements from CMS submit annually to CMS. Hospitals that are eligible for reimbursements from CMS must submit and certify a wealth of institution-specific cost information annually [9]. This submitted financial information is used by the regional Medicare Administrative Contractors (MAC) to calculate and publish the hospital-specific operating and capital CCRs via the impact files [10]. All 114 hospitals in our analysis were captured in the FY 2020 Impact File and were included in our financial analysis.

Methodology recommended by the Research Data Assistance Center (ResDAC), a CMS data contractor, was used to calculate hospital-level costs for admissions with rpFVIII and rFVIIa treatment by using hospital-level CMS CCRs for each of the 114 providers included in our analysis [11]. Hospital costs were calculated as:

$Hospitalcosts=CCR\ast thetotalchargedbyhospital(calculatedeachforrpFVIIIandrFVIIa)$

Once hospital costs were calculated, the difference or margin between hospital costs was determined by assessing the difference between hospital costs, as calculated above, and the amount reimbursed by CMS for each hospitalization. Therefore, margin was calculated as the difference between the CMS reimbursement and average hospital costs:

$FinancialImpact(orMargin)=Reimbursedamount-Hospitalcosts$

A positive margin is one where the difference between the reimbursed amount and average hospital costs is greater than zero; a negative margin is where this difference is less than zero. This methodology allowed us to determine the costs each hospital incurs for admissions with rpFVIII or rFVIIa treatment compared to the amount they are reimbursed for each admission to better understand the overall margin and financial impact for hospitals. To understand the overall impact of hospitalizations with rpFVIII or rFVIIa treatment, mean (SD) hospital costs, charges, and CCR were calculated for all hospitalizations in the study.

To evaluate the robustness of our findings, a sensitivity analysis was conducted using a range of CCRs for all 3,315 hospital providers in the FY 2020 Impact File, as the CCRs in the study hospitals may not have reflected the full set of CCRs across all hospitals captured in the Impact File. The interquartile range of CCRs from the FY 2020 Impact File was used to better understand CCRs beyond those in this study sample as a sensitivity analysis.

3. Results

There were a total of 220 hospitalizations with either rpFVIII or rFVIIa across 114 hospitals in our sample. Of these, 40 were treated with rpFVIII and 180 were treated with rFVIIa. As shown in Table 3, the mean hospital-wide cost-to-charge ratio across all admissions with either rpFVIII or rFVIIa was 0.26 (SD: 0.08). The mean charged amount was $1,588,176 (SD: 2,526,030), and the mean reimbursed amount was $394,751 (SD: 588,601). Consequently, the overall mean calculated margin was $38,394 across all admissions in our study. Average costs of hospitalizations were higher for hospitalizations with rpFVIII, compared to rFVIIa (rpFVIII: $452,092, rFVIIa: $335,082). However, the average reimbursed amounts were also slightly higher for patients treated with rpFVIII (rpFVIII: $503,641, rFVIIa: $371,026). Consequently, hospital margins were similar for admissions that used rpFVIII compared to rFVIIa (Mean: rpFVIII: $51,549; rFVIIa: $35,944).

Table 3. Cost outcomes for admissions with treatment.

Cost Outcomes	rpFVIII	rFVIIa	Overall Hospitalizations
	(n = 40)	(n = 180)	(n = 220)
	Mean
	(SD)
Reimbursed amount to hospital	$503,641	$371,026	$394,751
	($539,779)	($97,490)	($588,601)
Charged amount from hospital	$ 1,819,019	$1,536,877	$1,588,176
	($ 1,689,816)	($2,677,688)	($2,526,030)
Average Medicare Cost-to-Charge ratio (CCR)	0.27 (0.06)	0.26 (0.09)	0.26 (0.08)
Cost to hospital (Charged amount * CCR)	$ 452,092	$ 335,082	$ 356,357
	($369,503)	($ 469,103)	($ 454,119)
Impact/Margin (Reimbursed amount minus Cost to Hospital)	+$51,549	+$ 35,944	+$ 38,394

(+) indicate postive margins , i.e. hospital cost was less than the amount reimbursed to the hospital. Margins were calculated as reimbursed amount to hospital minus cost to hospital. (SD)=standard deviation

3.1. Sensitivity analysis

The mean CCR for all 3,315 facilities in the 2020 Impact File was 0.32 (SD: 0.16; Median (IQR): 0.29 (0.21, 0.39)), representing the national average, compared to a mean CCR of 0.27 (SD: 0.10; Median (IQR): 0.27 (0.20, 0.32)) for the 114 facilities in our sample. Table 4 displays the expected impact of rpFVIII on hospital finances for a range of CCR values from 0.20 to 0.50. These results suggest that the use of rpFVIII has an impact on hospital finances that is comparable to rFVIIa, with larger margins that are slightly more positive with rpFVIII at CCRs of 0.2 and 0.25, and negative margins that are smaller with rpFVIII at CCRs 0.3 through 0.45. At a CCR of 0.47, the margins are the same; at CCRs greater than 0.47, the negative margins are larger for rpFVIII. Therefore, impacts to hospital margins would not be expected with rpFVIII use unless their CCR exceeds 0.47.

Table 4. Sensitivity analysis for admissions with treatment.

Sensitivity	rpFVIII		rFVIIa
CCR	Hospital Cost*	Margin**	Hospital Cost	Margin**
0.2	$ 363,804	$139,837	$ 307,375	$ 63,650
0.25	$ 454,755	$48,886	$ 384,219	($13,193)
0.3	$ 545,706	($42,065)	$ 461,063	($90,037)
0.35	$ 636,657	($133,016)	$ 537,907	($166,881)
0.4	$ 727,607	($223,967)	$ 614,751	($243,725)
0.45	$ 818,558	($314,918)	$ 691,595	($320,569)
0.47	$ 854,939	($351,298)	$ 722,332	($351,306)
0.475	$ 864,034	($360,393)	$ 730,017	($358,991)
0.5	$ 909,509	($405,869)	$ 768,439	($397,413)

*Hospital cost = sensitivity CCR x average charged [rpFVIII: $1,819,019; rFVIIa: $1,536,877] (see Table 2).

**Margin = Average reimbursed amount [rpFVIII:503,641; rFVIIa:$371,026] (see Table 2) – Hospital Cost*. Values in parentheses indicate negative margins.

4. Discussion

Our analysis suggests that while there may be higher reimbursement for rpFVIII hospitalizations, the use of rpFVIII is associated with similar hospital margins compared to rFVIIa, despite rpFVIII’s higher upfront unit cost. The robustness of our results is further evidenced by the sensitivity analysis, which demonstrated that even up to a CCR of 0.47, hospitalizations with rpFVIII continued to have a negligible financial impact as compared to hospitalizations using rFVIIa. Importantly, a CCR this high is uncommon: of the 3,315 hospitals in the FY 2020 Impact File, fewer than 15% had a CCR higher than 0.47. In this study, we found an average hospital-wide CCR of 0.26, but previously published data suggests that intensive care units (ICUs), where many patients presenting with bleeds due to AHA are treated, have higher CCRs [12,13]. Childers et al. examined the 2015 fiscal year financial statements from 289 acute care hospitals in California and found a median hospital CCR of 0.25 and a median medical/surgical ICU CCR of 0.35 [12]. Yet, even at a CCR of 0.47, the financial impact on hospital margins of the two treatment options is very similar.

The data on real-world mortality/readmission benefit of rpFVIII and the absence of a financial impact in the hospitals included in this study may be useful findings for institutions considering having formulary consensus for approved use and/or treatment protocols for rpFVIII use in patients presenting with AHA [7]. Up-front costs of a treatment are an important consideration; however, hospitals may be able to reduce mortality and readmissions in this population, as suggested by Batt et al. This study builds on Batt et al., by suggesting that these improved clinical outcomes can be achieved without a financial impact to hospitals, and may be used to help pharmacy and therapeutics committees make informed decisions about formulary and product stocking [7].

This study has several strengths. It provides much-needed insight into the financial impact to hospitals considering the expansion of their formularies to include rpFVIII for the treatment of AHA. We were able to use CMS CCRs for each of the 114 facilities in which patients in our sample received treatments with rpFVIII and/or rFVIIa. Our analysis used hospital-specific cost-to-charge ratios from CMS, as well as charged and reimbursed amounts from Medicare claims, to provide a complete, in-depth understanding of the financial impact on hospital margins from the use of these treatments in the real world. The sensitivity analysis provided additional insight into the potential margins resulting from using rpFVIII versus rFVIIa for hospitals nationally, allowing us to conclude that, for the vast majority of hospitals, we would not expect a financial impact associated with the use of rpFVIII, compared with rpFVIIa, for the treatment of AHA. However, variables such as clinical benefits, CPI, case-mix index (CMI), payer contracts, as well as acquisition prices would need to be considered on a case-by-case basis when assessing a therapy’s true value. Nonetheless, hospital formulary decision-making based solely on acquisition price may not allow for optimal financial balance to be reached.

Although this analysis calculated hospital costs from the charged amounts, providing robustness to the study, differences in DRGs and additional finances related to higher increased payments/reimbursement were out of scope for the current analysis, and thus not assessed in detail. Moreover, reimbursement rules are complex and influenced by many factors, independent of administered medications, such as outlier and other payment rules stemming from case mix, complications and comorbidities, and acuity of care. Of note, rpFVIII patients had a higher observed comorbidly burden (Table 2) and in some cases included the administration of both rpFVIII and rFVIIa as rpFVIII may not be immediately available [7]. Additionally, the ability to match on patient clinical characteristics was not possible due to the study sample size, limiting generalizability. Another limitation is the study’s reliance on patients with fee-for-service Medicare coverage, the large majority of whom are aged 65 years and above. Consequently, our study results may not be generalizable to commercially insured individuals with AHA. However [14–16], AHA primarily affects older patients and is most common in individuals 60–80 years. Lastly, this analysis should be repeated controlling for case mix, provider region and/or matching on providers when a larger dataset of those treated with each therapy of interest is available.

Article highlights

• Currently, bypassing agents (rFVIIa, off-label aPCC [anti-inhibitor coagulant complex]) and rpFVIII are accepted as treatments for patients with AHA presenting for bleed events.

• Some hospitals do not stock rpFVIII or allow use due to its higher per unit cost compared to rFVIIa; decision-makers should consider a broad clinical and financial view when considering treatments on formulary, as decision-making based solely acquisition price may not allow for optimal financial balance to be reached.

• This study used Centers for Medicare and Medicaid (CMS) cost-to-charge ratios, derived from real-world data from CMS, to assess the financial impact on hospitals of the use of rpFVIII and rFVIIa for the treatment of AHA.

• While there may be higher reimbursement in rpFVIII group, this analysis finds no evidence of an impact on hospital finances when utilizing rpFVIII compared to rFVIIa.

• Sensitivity analyses showed that negative impacts on hospitals’ financial margins are smaller with the use of rpFVIII compared to rFVIIa until the CCR exceeds 0.47, which occurs in fewer than 15% of hospitals.

Author contributions

All authors were involved in the conception and design of the study, analysis, or interpretation of the data, and drafting of the paper or revising it critically for intellectual content. All authors approved of the final version for publication and are accountable for all aspects of the work.

Declaration of interests

M Bullano and B Schultz are employed by Takeda with stocks in the company. C Cool, S Durgapal, N Sacks, Y Liu, R Kee, and K Batt received consultancy fees from Takeda. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethical conduct of research disclosure

The Medicare LDS data and Fiscal Year (FY) 2020 Impact Files from CMS used in this study did not contain data elements that would permit the identification of beneficiaries (patients) in accordance with privacy requirements of the Health Insurance Portability and Accountability Act (HIPAA Privacy Rule). Because of this, institutional review board approval for this research was not required.

Acknowledgments

Writing and editorial assistance for the development of this manuscript was provided by Jennifer Hernandez and Zachary Heim, who are employees of Precision Value and Health.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14737167.2023.2160705

Additional information

Funding

This study and manuscript writing was sponsored by Takeda Pharmaceuticals USA, Inc. Lexington, Massachusetts, USA.