Impact of pharmacy services on time to elexacaftor-tezacaftor-ivacaftor initiation

BACKGROUND: The approval of elexacaftor-tezacaftor-ivacaftor (ELX/TEZ/IVA) expanded highly effective cystic fibrosis transmembrane receptor modulator therapy to approximately 90% of persons aged 12 years and older with cystic fibrosis. Clinical pharmacists and pharmacy technicians played a key role in planning for ELX/TEZ/IVA initiation prior to US Food and Drug Administration approval as well as initiating therapy after approval. OBJECTIVE: To evaluate the impact of pharmacy services on time to ELX/TEZ/IVA initiation. METHODS: A retrospective chart review evaluated 146 patients aged at least 12 years with cystic fibrosis qualifying for ELX/TEZ/IVA at a single health system between October 21, 2019, and April 1, 2020. RESULTS: Patients filling ELX/TEZ/IVA at an integrated health system specialty pharmacy (HSSP) vs an outside specialty pharmacy (SP) started on therapy an average of 10.8 days sooner (10.8 days ± 14.0 vs 21.6 days ± 18.8, respectively; P = 0.006). More patients filling at an HSSP received ELX/TEZ/IVA within 14 days of the prescription being written compared with outside SPs (82.0% vs 41.4%, respectively; P = 0.001). Before ELX/TEZ/IVA initiation, patients were hospitalized for a cystic fibrosis–related complication for an average of 6.26 days (range = 0-183) compared with 1.16 days (range = 0-91) after ELX/TEZ/IVA initiation. Lastly, an estimated $134,810 was saved in hospitalization dollars in the 105 patients that were able to fill ELX/TEZ/IVA at an HSSP by initiating the drug an average of 10.8 days sooner than outside SPs. CONCLUSIONS: The results of this study demonstrate the value of an integrated HSSP model. The ability to fill specialty medications at an integrated HSSP may optimize medication access, control costs, and improve patient outcomes for patients receiving care within a health system.

Cystic fibrosis (CF) is a progressive genetic disorder caused by mutations in the CF transmembrane conductance regulator (CFTR) protein. These mutations lead to fluid dysregulation in various organ systems resulting in mucus buildup and subsequent complications. The most common complication of CF is persistent lung infections, which can result in frequent hospitalizations and diminished pulmonary function. 1 The standard of care for CF treatment now includes a class of medications called CFTR modulators (CFTRms). 2 The commercially available CFTRms use 2 distinct mechanisms, classified as "potentiators" or "correctors." 2 Potentiators exert their effect by opening the CFTR protein gate to allow chloride through the cell membrane and correctors assist in the proper formation of CFTR proteins to enable trafficking to the cell surface. 2 Together, the 2 effects can improve the function of the CFTR protein, therefore improving the function of multiple organ systems through the regulation of chloride concentrations. 1 The development of CFTRms has allowed for improvements in lung function, reduced rates of disease progression, fewer pulmonary exacerbations and hospitalizations, increased nutritional status, and enhanced quality of life for patients with CF who qualify for the therapy. 2 Elexacaftor-tezacaftor-ivacaftor (ELX/TEZ/IVA) is a novel CFTRm with a unique mechanism as it contains 2 correctors, elexacaftor and tezacaftor, along with a potentiator, ivacaftor. ELX/TEZ/IVA's benefits include improved lung function, reduced rates of pulmonary exacerbations, lower sweat chloride concentrations, higher CF Questionnaire-Revised respiratory domain scores, and increased body mass index scores with minimal adverse effects. 3 These benefits led to its approval in October of 2019. 3 At the time of this study, it was approved for the use in patients with CF at least 12 years of age who have one copy of the F508del CFTR mutation 4 and was the first modulator approved for patients with a single F508del CFTR mutation. Since study completion, ELX/TEZ/IVA approval was extended to children with CF who were aged 6 through 11 years and additional rare mutations for ages 6 years and older. 5,6 The approval of ELX/TEZ/IVA was monumental as it expanded the profound benefits of CFTRm to a group of patients who previously did not qualify for highly effective treatment with CFTRms from approximately 10% with ivacaftor alone to more than 90% with ELX/TEZ/IVA. 4,7 This retrospective review took place at a large, single academic health system encompassing several hospitals, ambulatory clinics, and retail pharmacies including a specialty pharmacy (SP). The health system has become one of the largest adult CF care centers in the Midwest region, caring for approximately 270 adults and approximately 50 pediatric patients with CF. The institution has 8 specialized CF providers and a comprehensive support team consisting of clinical pharmacists, a pharmacy technician, nurses, social workers, a dietician, a psychologist, and respiratory therapists. The health system's integrated health system SP (HSSP) model, similar to the "integrated pharmacy team" described by Zobell et al, 8 encompasses the constant interaction between patients, pharmacy patient advocates (PPAs), ambulatory clinical pharmacists, and the SP. The PPA role is filled by certified pharmacy technicians and is similar to the SP technician role described by Zobell et al. 8 The integrated HSSP model supports patients by determining benefits eligibility, completing prior authorizations (PAs) and appeals for insurance approval, and obtaining financial assistance through assistance programs such as copay assistance cards or grants. This model has been shown to improve medication access, adherence, appropriate medication usage, and communication with providers or other health care members within a health system. [8][9][10][11] HSSP models have shifted the focus to providing efficient and safe patient care rather than on focusing primarily on specialty medication dispensing. 12 Previous studies have shown the benefit of pharmacy services on improving CF medication adherence and patient outcomes [12][13][14][15][16][17] as well as the benefit of an integrated pharmacy team and SP on decreasing the time to insurance approval 18 and the time to the initiation of treatment (turnaround time). 19,20 The purpose of this study was to evaluate the impact of pharmacy services on time to ELX/ TEZ/IVA initiation with a focus on turnaround time from an HSSP compared with outside SPs.

Methods
A retrospective, single-center chart review included patients 12 years of age or older with at least one F508del mutation in the CFTR gene prescribed and initiated on ELX/TEZ/IVA between October 21, 2019 (date of US Food and Drug Administration [FDA] approval for ELX/TEZ/ IVA), and April 1, 2020. Data on these patients were collected from July 1, 2018, to October 1, 2020. For this study, initiation was defined as the date the medication was dispensed from the SP. Patients were excluded if they were enrolled in the ELX/TEZ/IVA study 3 or early access program, received free drug from the manufacturer, underwent lung transplantation prior to drug initiation, or did not initiate the drug in the above time frame. Additionally, patients were excluded from hospitalization and survival analysis if they had less than 6 months of follow-up available before and accurate fill data for all patients. 21 Turnaround time was defined as the time from the date electronically prescribed within the electronic health record to the date dispensed from the SP.
Categorical data were assessed with chi-square tests of Fisher's exact tests, and continuous data were analyzed with Student's t-tests where appropriate. Median values were assessed using the Mann-Whitney U test. Results were considered statically significant at P < 0.05 using a 95% CI. This study was approved by the University of Kansas Medical Center Institutional Review Board.

Results
Of the 201 patients evaluated, 146 were included in the pharmacy analysis portion of the study (Figure 2). An additional 13 patients did not have 6 months' follow-up either before or after ELX/TEZ/IVA initiation and were excluded from the hospitalization and survival analysis. The most common reason for exclusion was enrollment in the early access program through the drug manufacturer ( Figure 2). or after ELX/TEZ/IVA initiation. Six months was chosen as the minimum follow-up period to mirror the ELX/TEZ/ IVA landmark approval study. 3 After the initial FDA approval of ELX/TEZ/IVA, the pharmacy team played a key role in the medication initiation process within the adult and pediatric CF clinics. The clinical pharmacists and PPAs worked with the multidisciplinary team to ensure efficient and successful drug initiation for all eligible patients receiving care within the health system. The workflow for ELX/TEZ/IVA initiation is described in Figure 1.
Medication dispense data, including medication dispense dates, for patients filling at the HSSP was accessed from the prescription dispensing system. Prescription dispense data for patients filling at outside SPs was obtained from the CFTRm manufacturers Guidance and Patient Support (GPS) program. 21 All patients filling ELX/TEZ/IVA at an outside SP were enrolled in GPS after giving signed or verbal consent. Pharmacies with access to purchase and dispense the medication are required to share real-time dispense data for any patient enrolled in the GPS program, allowing for complete compared with those filling at outside SPs (10.8 ± 14.0 days vs 21.6 days ± 18.8 days, respectively; P = 0.006; Figure 3). Similarly, the average time from insurance approval to dispense was 11.9 days quicker with the HSSP compared with outside SPs (4.8 ± 6.5 days vs 16.7 ± 18.7 days; P = 0.002; Figure 3). Additionally, significantly more patients received ELX/TEZ/IVA within 14 days of the date prescribed if they used the HSSP compared with outside SPs (82.0% vs 41.4%, respectively; P = 0.001). Figure 4 further demonstrates the overall improvement in the time to ELX/TEX/IVA initiation when filling at the HSSP compared with outside SPs. A PA was required and completed on 138 patients (94.5%). On average, PAs were completed within 1.3 (± 2.1) days of being prescribed and were approved in 5.5 (± 1.5) days after submission. There was no difference in the time to PA response between payor type (median 2 days commercial vs 1 day government). Upon PA denial, an initial appeal was submitted to the insurance with a classification of urgent for all 23 PAs that were denied, and a response was received Overall, baseline characteristics were similar among patients who filled at the HSSP compared with outside SPs with the exception of baseline forced expiratory volume in the first second of expiration percentage (FEV 1 %) and the number of patients initiated on ELX/TEZ/IVA during an inpatient hospitalization (Table 1). Regarding third-party coverage details, the majority of patients filling at outside SPs (96.6%) had commercial insurance compared with only 62.3% of patients filling at the HSSP. All patients with government insurance (Medicare and Medicaid) filled at the HSSP (Table 1). When further evaluating FEV 1 % in correlation with third-party coverage, 45.1% of patients with government plans had a FEV 1 % of 50% or less in comparison with 23.2% of patients with commercial or Tricare insurance plans.
Overall, 65.8% (n = 96) of patients were initiated on ELX/ TEZ/IVA within 60 days of FDA approval and 83.6% (n = 122) within 90 days. Patients filling ELX/TEZ/IVA at the HSSP had an average turnaround time that was 10.8 days faster approval. Patients filling at the HSSP experienced a significantly shorter time to ELX/TEZ/IVA initiation compared with those required to fill at outside SPs. Additionally, patients also incurred a reduction in the number of hospitalization days because of CF-related complications after starting ELX/TEZ/IVA. To our knowledge, this is the first study analyzing the impact of pharmacy services on the time to ELX/TEZ/IVA initiation after FDA approval.
Our study found that patients started on ELX/TEZ/IVA 10.8 days sooner when filling with our HSSP compared with those filling at an outside SP. Several other studies have shown a similar reduction in turnaround time or time to initiation of therapy with the HSSP model. 11,[18][19][20] A study of patients with CF prescribed inhaled dornase alfa found a decreased use of multiple pharmacies and an improved time to delivery with pharmacy support. 19 Wright and colleagues use an HSSP model for the initiation of CFTRms in pediatric patients with CF and demonstrated a turnaround time of 6.8 ± 8.6 days at the HSSP compared with 15.4 ± 15.1 days at an outside SP (P = 0.005). 18 Our turnaround time was slightly longer at 10.8 ± 14.0 days compared with their 6.8 days. This is likely because of our focus on the initiation of ELX/TEZ/IVA immediately upon approval from the FDA and delays with insurers adding the drug to formulary after drug approval. Both of the above studies demonstrated the value of an HSSP model on improving turnaround time and allowing patients to start therapy sooner; however, neither evaluated this after initial FDA approval. We showed a reduction in the time to initiation of more than 50% if the drug was filled within an HSSP compared with an outside SP. This was crucial in allowing a large group of patients to start an effective treatment for the first time since their 391 days evaluated before ELX/TEZ/ IVA initiation, patients were hospitalized for a CF-related complication for an average of 6.26 days (range = 0-183). An average of 292 days per patient were evaluated after ELX/TEZ/IVA initiation through the end of the study period: October 1, 2020. During the 292 days evaluated after ELX/TEZ/IVA initiation, patients were hospitalized because of a CF-related complication an average of 1.16 days (range = 0-91). At the patient level, there was a mean decrease of 0.047 (± 0.079) hospitalization days normalized per patient day. Lastly, only 1 patient underwent lung transplantation 201 days after ELX/ TEZ/IVA initiation, which ultimately resulted in death.

Discussion
This study demonstrated a high rate of success in achieving insurance approval for ELZ/TEZ/IVA for a large group of patients meeting the criteria and starting therapy following FDA an average of 4.6 (± 6.6) days later. The initial appeal was denied for 5 patients. All 5 second appeals were approved with an average time to approval of 6.0 (± 5.6) days. This process and support from the internal HSSP team was the same for all patients, regardless of which SP they used for dispensing ( Figure 1).
The average final out-of-pocket cost for patients filling ELX/TEZ/IVA at the HSSP was $3.55 (± $10.6). Prior to obtaining copay assistance, the average insurance copay was $245.39 (± $878.54), resulting in an average outof-pocket savings from HSSP efforts of $241.84 per patient. Of the 40 patients filling at the HSSP and receiving copay assistance, 27.5% used a manufacturer copay card, 32.5% used a grant, and 40.0% used both.
A total of 133 patients were included in the final hospitalization and survival analysis (Figure 2). An average of 391 days per patient was evaluated for hospitalization data before ELX/TEZ/ IVA initiation within the time frame of July 1, 2018, to July 31, 2019. During the

TABLE 1
Baseline Patient Characteristics to have government insurance (43.6%) compared with those filling outside (0%; P < 0.001). Government payors are required to allow patients to fill at the pharmacy of their choice. All of our patients with government insurance chose to fill with our HSSP. As a result, they were able to start on the drug sooner, which could have resulted in a decreased use of hospital days. Furthermore, our hospitalization analysis suggests this delay may contribute to worse outcomes in terms of hospitalization days, the associated financial impact, and morbidity related to the hospitalization itself. Upon review of baseline demographics, it was seen that patients filling at the HSSP had a significantly lower average baseline FEV 1 % compared with patients filling at outside SPs. Additionally, almost half of the patients with government insurance had a FEV 1 % of 50% or less compared with approximately one-quarter of patients with commercial or Tricare insurance plans. The difference in baseline lung function between the groups is likely due to the social security disability benefit qualifications for patients with CF. The United States Social Security disability criteria for patients with CF are based on several factors, including FEV 1 volume. 26 The FEV 1 volume criteria vary by age, sex, and height but equate to a FEV 1 % of approximately 50% or less. Patients that receive social security disability benefits may also qualify for government insurance. Our finding that patients with government insurance had a lower FEV 1 % than patients with commercial plans correlates with the United States Social Security disability criteria for CF. 26

LIMITATIONS
There are limitations to consider, including those inherent to a retrospective study. Specifically, progress notes were used to document and record most timepoints in the drug approval process, including PA submission, appeal submission, insurance approval, and education. Although standard CF diagnosis because of the more inclusive qualification criteria for ELX/TEZ/IVA.
Hospitalizations can cause significant financial burden in patients with CF. A cost analysis was performed to evaluate health care savings associated with the early initiation of ELX/TEZ/IVA and subsequent reduced hospitalization days for patients filling at the HSSP. Based on initiating the drug an average of 10.8 days quicker when filling at the HSSP, patients filling at our HSSP had 0.047 fewer hospitalization days normalized per patient day after ELX/ TEZ/IVA initiation. Using the average cost of 1 adult CF hospitalization day ($2,529.39), 22 an estimated $134,810 of hospitalization day costs were saved in the 105 patients that were able to fill ELX/TEZ/IVA within the HSSP. In addition to the cost savings, by initiating therapy quicker and reducing the hospitalization days caused by CF-related complications, patients are less susceptible to further complications, including worsening structural lung changes, poorer nutritional outcomes, reduced quality of life, and increased financial burden. 23,24 In our study, patients filling at an outside SP were more likely to have commercial insurance (96.6%) compared with those filling at the HSSP (53.8%; P < 0.001). Many patients with commercial insurance plans are mandated by their insurance to fill specialty medications at an outside SP rather than within an HSSP because of payor network restrictions, 25 likely contributing to the higher number of patients with commercial insurance filling outside. Moreover, although the pharmacy team provided the same level of support to patients initiating ELX/TEZ/IVA, regardless of filling pharmacy, patients filling at an outside SP had a longer time to ELX/TEZ/IVA initiation. Our study suggested that insurance payer mandates requiring patients to fill at certain SPs led to delays in treatment initiation. Additionally, patients filling at our HSSP were more likely evaluated before initiation and 292 days per patient after initiation, the number of hospitalizations after initiation could be underestimated and the decrease in hospitalization days may be overestimated. Furthermore, although our patient population is rarely admitted outside of our health system, we did not have data for hospitalizations that may have occurred outside. Additionally, the landmark trial for ELX/TEZ/IVA used hospitalizations per patient-year to describe their findings regarding reduced hospitalization rates. 3 Unfortunately, we cannot compare our reduction in hospitalization days with the landmark trial because we used an alternative unit of measure. 3 Further exploration is warranted to confirm if a similar incidence of hospitalization reduction was found in our patient population compared with the approval study population.

Conclusions
These findings further validate the improved outcomes patients incur with the HSSP model. Despite patients receiving the same pharmacy support, workflows were in place, the process is dependent on the user completing documentation and doing so in a timely fashion that is not without potential flaws, though this would be similar between groups. Additionally, unforeseen delays in the approval process or drug initiation were not accounted for and could have impacted our results. However, these delays would have potentially overestimated the time to drug initiation for patients in both groups, would be expected to be similar between groups, and, if accounted for, would have likely strengthened the results. Another limitation is the variability in hospitalization timeframes evaluated before and after ELX/TEZ/ IVA initiation. The exclusion criteria ensured a minimum of 6 months of follow-up both before and after drug initiation; however, given that the nature of the various start dates of ELX/TEZ/IVA, we were not able to standardize the average of days evaluated before and after drug initiation. Although we saw a decrease of 0.047 (± 0.079) hospitalization days normalized per patient day, because an average of 391 days per patient was

FIGURE 4
Overall Time to ELZ/TEZ/IVA Initiation