Evaluation of Dual Versus Triple Therapy for Hepatitis C
In this retrospective chart review, patients with hepatitis C had a sustained virologic response with either dual or triple therapy.
Chronic hepatitis C virus (HCV) is estimated to affect more than 150 million people worldwide, exceeding human immunodeficiency virus (HIV) as the leading cause of blood-borne pathogen-induced mortality in the United States.1 It is responsible for substantial morbidity, including chronic liver disease, cirrhosis, and hepatocellular carcinoma that may require liver transplantation.2 Eleven major genotypes of HCV have been identified, with genotype 1 being the most common in the United States and the hardest to treat.3 The goal of treatment is eradication of the virus to prevent complications associated with the disease4 and decrease all-cause mortality.5 Eradication is confirmed by the measurement of sustained virologic response (SVR) defined as the absence of HCV RNA in blood 6 months after hepatitis C therapy is complete. Until 2011, the American Association for the Study of Liver Disease guidelines recommended dual therapy, a combination of pegylated interferon and ribavirin, to treat chronic HCV infection. This treatment regimen produces sustained virologic response in 40% to 50% of patients with HCV genotype 1.4-6 After FDA approval of the protease inhibitors telaprevir and boceprevir, practice guidelines were updated to include the addition of either agent to dual therapy for treatment of patients with HCV genotype 1 (triple therapy).4 The addition of these agents was expected to increase SVR by approximately 20% and shorten drug therapy for some patients.
Increased efficacy with these agents has resulted in a tradeoff with increased side effect profiles and toxicities.7-14 Management of these additional side effects has resulted in new unintended expenses such as extra hepatitis C clinical nursing resources, physician visits, specialty consultations, hospital admissions, emergency department (ED) costs, transfusions, more frequent laboratory analysis, and additional concomitant medications. These extra expenses are incurred by patients, health plans, clinics, hospitals, and the medical system at large.
Prior studies have examined the financial impact of HCV and have found that nearly one-third of the costs incurred with treatment of the virus are directly related to the disease itself, with one-half of all drug costs associated with HCV-specific treatment.15 Another study found that patients who are undergoing therapy for HCV have twice the mean number of ED visits, hospitalizations, and traditional provider visits than control groups.16 As our nation reflects on the sticker shock of the newest direct antiviral drugs that have been approved, more careful evaluation of the total cost of therapy and realistic expectations about associated costs are imperative. This will require more knowledge than simply SVR and cost per pill. Safety and side effects with their associated costs must be fully analyzed.
The purpose of this study was to compare traditional dual-therapy regimens (pegylated interferon and ribavirin) with triple-therapy regimens (pegylated interferon and ribavirin plus either boceprevir or telaprevir) with respect to patient outcomes, including evaluation of efficacy, safety, and side effect outcomes. We hypothesized that triple-therapy regimens would increase overall SVR in treatment-naïve genotype 1 HCV patients; however, this increase would come with an additional increase in side effects, number of hospital admissions, ED visits, transfusions, and concomitant medications in this cohort group.
We conducted a retrospective chart review comparing efficacy (SVR), safety (number of hospitalizations, ED visits, transfusions), and side effect outcomes (patient-reported side effects, treatments, consultations, additional medications, dose reductions) in treatment-naïve chronic HCV genotype 1 patients treated with either dual or triple therapy. Data were collected from electronic and written medical records of approximately 300 patients initiated on HCV treatment by the Division of Gastroenterology, Scott & White Healthcare (SWHC), between July 1, 2004, and December 31, 2012. Approval was received from the Scott & White Institutional Review Board before data collection.
Setting and Participants
Eligible patients were initially identified from the SWHC data warehouse with International Classification of Diseases, Ninth Revision (ICD-9) code 070.54 (chronic hepatitis C without mention of hepatic coma) or 070.70 (unspecified viral hepatitis C without hepatic coma). These patients were prescribed pegylated interferon and ribavirin with or without boceprevir or telaprevir. This database was cross-matched with patients seen in the SWHC Hepatitis C Clinic for HCV therapy. Patients were considered for inclusion if they were over 18 years of age, positive for HCV genotype 1a or 1b, treatment-naïve, and treated by SWHC on an outpatient basis. We excluded persons who were prior transplant patients with recurrent HCV, were co-infected with HIV/acquired immunodeficiency syndrome or hepatitis B, were receiving hemodialysis, were diagnosed with acute hepatitis C without diagnosis of chronic HCV, or had other contraindications to HCV treatment including autoimmune hepatitis, hemoglobinopathies, renal dysfunction, or pregnancy. Patients were assigned to either dual-therapy or triple-therapy cohorts based on initial treatment received (
Patients were seen at least monthly throughout the entire course of their therapy by either a dedicated hepatitis C nurse and/or a treating physician. All of the patients were telephoned weekly by a hepatitis C nurse who confirmed compliance with treatment and assessed for any new side effects. All of the visits and telephone communications were recorded in both the electronic medical record (EMR) and the patient’s hepatitis C written flow sheet. The flow sheet recorded all laboratory, dosage changes, new medications, and other issues that the patient experienced during therapy.
During retrospective chart review, data were collected on the intended treatment plan from clinic EMR notes and written flow sheets, and on the outcome with respect to SVR from laboratory data, flow sheets, and the EMR. Adverse events such as hospitalizations, ED visits, and transfusions were verified from the EMR, flow sheets, and interviews with the hepatitis C nurses. Side effect occurrence, management (including new drugs given, consultations placed, transfusions, dose reductions, and so forth), and outcomes were assessed from the EMR, flow sheet review, and interview with the hepatitis C nurses. These events were considered treatment associated if they occurred after the first dose of hepatitis drug therapy and up to 3 months after the last dose of drug therapy. If there was any question of association, the treating physician was asked to render an opinion.
Primary outcomes were the percentage of patients who achieved SVR (defined as an undetectable HCV RNA level 6 months after last dose of medication) and the percentage of hospitalizations, ED visits, and transfusions during the patient treatment period. Other outcomes included incidences of common side effects and respective treatment strategies used to mitigate these side effects. Additionally, a subanalysis of medication adherence was conducted on the data of those patients covered by the SWHC, utilizing electronic fill data. Adherence was determined through calculation of the medication possession ratio as days of medication supplied divided by total treatment interval.17 The total treatment interval was the total number of days the patient was to be on therapy. Percentage of compliance was the proportion of those patients with a medication possession ratio greater than 0.8.
We used SAS version 9.2 (SAS Institute, Cary, North Carolina), for data analysis. A P value of less than .05 was considered to be statistically significant. Sample size calculations were based on SVR achieved in published randomized controlled trials (40% for dual therapy, 70% for triple therapy). It was estimated a total of 100 patients would be required to have 80% power to detect a 30% difference in SVR between our treatment cohorts (http://biomath.info/power/chsq.htm).
Baseline characteristics are presented by treatment cohort as mean and standard deviation for continuous variables and number of observations with percentages for categorical variables. P values were determined with a χ2 test without Yates correction factor for categorical variables and a t test for continuous variables.
A total of 337 patients were identified. Of these, 208 patients did not meet inclusion criteria and 24 were excluded. A total of 105 patients were analyzed, with 81 in the dual-therapy cohort and 24 in the triple-therapy cohort (Figure). All baseline characteristics were similar between patient cohorts with exception of HCV RNA, which was significantly higher in the triple-therapy cohort (P <.001; Table 1).
Efficacy and Safety
The efficacy outcome of SVR was significantly higher in the triple-therapy cohort versus the dual-therapy cohort (67% vs 42%, P = .033) (
). Triple-therapy SVR was directly related to the ability of the patients to complete intended therapy. Of the 14 patients who did not meet stopping rules and did not stop early because of side effects, 11 achieved SVR and the remaining 3 were lost to follow-up. None of the triple-therapy patients who completed 12 weeks or fewer of therapy achieved SVR. For the safety outcomes there were more hospitalizations (12.5% vs 11.1%, P = .851), ED visits (25% vs 21%, P = .676), and transfusions (8.3% vs 0%, P = .009) in the triple-therapy cohort versus the dual-therapy cohort, although only 1 of these was statistically significant.
Side Effect Prevalence
Overall, 580 side effects were reported (
). All but 3 patients experienced at least 1 side effect. The most prevalent side effects for both cohorts included fatigue (66%), anxiety/depression (47%), anemia (35%), nausea (32%), and rash (25%).
Patients on triple therapy experienced more side effects than patients on double therapy; however, most of these effects did not achieve statistical significance. Exceptions were anemia (67% vs 26%, P = .0002), nausea (63% vs 23%, P = .003), and shortness of breath (25% vs 9%, P = .0045). Side effects seen in a greater percentage of dual-therapy patients included abdominal pain, chills/fever, neutropenia, oral ulcers, pain not specified, short-term memory loss, thrombocytopenia, and weight loss, none of which reached statistical significance.
Side Effect Management
New Prescriptions. This study also examined the outcomes of side effects and which strategies were used to mitigate these effects (Table 4). Side effect outcomes were recorded for patients experiencing at least 1 side effect. Patients may have experienced multiple side effects, may have received more than 1 medication, and may have experienced a side effect that required multiple management strategies (eg, anemia may have been treated with a colony-stimulating factor, but later required the patient to stop all HCV treatment). The majority (76%) of patients were prescribed at least 1 medication for treatment of a reported side effect, for a total of 90 new prescriptions in 105 patients. In both cohorts, 86% of patients used extra nonprescription medications to sustain themselves through therapy.
A total of 33 patients were given colony-stimulating factors. This was more common in the triple-therapy cohort than in the dual-therapy cohort (46% vs 27%, P = .0835). Anemia, the most commonly treated side effect, was treated with epoetin (n = 25), followed by neutropenia treated with filgrastim (n = 6) and thrombocytopenia treated with eltrompobag (n = 2). One patient in the dual-therapy cohort also received oprelvekin for thrombocytopenia. In the triple-therapy cohort, epoetin was given an average of 8 doses of 40,000 units a week (range: 2-28 doses).
Dose Adjustments. Patients were much more likely to receive a dose reduction of HCV treatment medications if they were in the triple-therapy cohort (54% vs 10%, P <.001). A total of 25 dosage reductions were recorded for 21 patients (13 of 24 in the triple-therapy cohort vs 8 of 81 in the dual-therapy cohort). The most common dosage change was a reduction in ribavirin for anemia (11 in triple therapy, 5 in dual therapy). Dosage reductions for interferon (4 in triple therapy, 5 in dual therapy) were for anemia (n = 1), anxiety/depression (n = 1), fatigue (n = 1), neutropenia (n = 2), weight loss (n = 2), and thrombocytopenia (n = 2).
Treatment Drug Discontinuation. A total of 18 patients from both cohorts discontinued all HCV treatment because of side effects. Of the 24 patients in the triple-therapy cohort, 5 experienced side effects requiring discontinuation of their protease inhibitor (with continuation on dual therapy alone); 3 patients experienced rash, 1 experienced rash with drug rash with eosinophilia and systemic symptoms (DRESS), and 1 patient experienced anemia. Only 1 patient in the triple-therapy cohort discontinued all treatment; the discontinuation was secondary to alopecia. The remaining 17 discontinuations were originally in the dual-therapy cohort. Side effects leading to treatment cessation included psychological issues (anxiety/depression = 7; suicidal ideation = 2; paranoia = 1), cardiovascular issues (stroke = 1; ruptured coronary vessel = 1); blood dyscrasias (neutropenia = 2; thrombocytopenia = 3), and flu-like symptoms (n = 1).
Adherence. Eighty-three percent of patients completed their intended HCV therapy. Treatment regimen adherence was analyzed from patients participating in the Scott & White Health Plan. Compliance in the dual-therapy cohort was 80% for peginterferon and 78% for ribavirin. In the triple-therapy cohort, overall compliance was 100% for peginterferon, 60% for ribavirin, and 80% for the protease inhibitor.
Despite triple-therapy regimens being shorter, the intensity of monitoring for side effects and utilization of clinical resources in the hepatitis C clinic was approximately double that of traditional dual-therapy patients. Before triple-therapy initiation, much extra effort was spent to maximize compliance (because these drugs required dosing 3 times a day for 3 to 12 months and were known to have worse side effects). Patients and people in their support system were required to attend extra pretreatment clinic visits, education classes, and one-on-one counseling with the hepatitis C nurse to ensure compliance and prepare for arrival of potentially life-threatening side effects. Only the best candidates for triple therapy were ever initiated on this demanding protocol. During therapy, these patients required double the usual number of blood draws to monitor for rapidly progressing anemia. They also required immediate clinical appointments for providers to examine any new rashes the same day that the rash appeared. The frequent dosage adjustments and additions of expensive and difficult to obtain and administer colony-stimulating factors also led to a large amount of resource utilization (clerical, nursing, provider, pharmaceutical, preauthorization, co-pay, patient inconvenience, laboratory, frequent callbacks, and patient anxiety).
Results from this retrospective cohort study represent a real-world experience with HCV therapy in central Texas. Previous randomized controlled trials have shown 40% to 50% SVR for dual therapy.4 For triple therapy, SVR response for telaprevir has been shown to be 75% to 92%9,10 and 63% to 66%13 for boceprevir. Our study showed comparable SVR rates of 42% for dual therapy and 66.7% for triple therapy.
Limitations of this study are related to the retrospective trial design. This study was heavily dependent on the EMR, written flow sheet documentation, and dedicated hepatitis C nurses. Most information was gathered from progress notes dictated by physicians and the full-time hepatitis C registered nurse, and some inferences had to be made on occasion when inconsistencies were found.
Another limitation is the large difference in sample size between the cohort groups. The triple-therapy cohort was too small for the boceprevir telaprevir data to be separated. Of 6 patients who did not have 6-month SVRs recorded, 4 were in the triple-therapy group. All of these patients had undetectable HCV at 3 months after therapy as well as normal liver function tests. Although our sample size was small, our primary outcome of SVR was statistically significant.
In conclusion, HCV patients treated at SWHC had an SVR with both therapy regimens that approach the level of national trials. Side effects were similar in both cohorts, with the exception of statistically higher rates of anemia, nausea, and shortness of breath in the triple-therapy cohort. In response to side effects in patients on triple therapy, 46% were prescribed a colony-stimulating factor, 54% had at least 1 dose reduction, 8.3% received a blood transfusion, 88% had at least 1 additional prescription written to sustain them through their side effect(s), and 1 patient discontinued therapy because of the ominous DRESS syndrome. Triple therapy resulted in more resources utilized to deal with more side effects, but resulted in the expected higher SVR of 66.7% compared with 42% in the dual-therapy cohort. Medication compliance was high for both cohorts: 80% to 100% for peginterferon, 60% to 78% for ribavirin, and 80% for the protease inhibitors. Because the CDC has recommended 100% screening of baby boomers for HCV18 and it is estimated that 800,000 new cases of HCV will be diagnosed, it is crucial that modern healthcare providers closely scrutinize new therapies that eradicate this virus. Some of the newest therapies are projected to cost close to $100,000 for the medications alone. If these new regimens deliver a 90% to 100% SVR after several weeks of all-oral treatment with very few side effects, this might be a reasonable price to pay. Also, many of the new therapeutic compounds have a goal of the least pill burden for the least number of days on therapy. Triple therapy—with the high burden of taking a pill every 8 hours for 3 to 12 months in addition to ribavirin pills taken twice a day and painful interferon injections once a week—was met with surprisingly high compliance. Therefore, with a strong support program in place, new drug regimens should continue to focus on the SVR rate, side effect profile, and the cost as being more important to patients than perceived pill burden.
We are completing a study quantifying the actual dollars spend on treatment of the side effects associated with triple therapy. It is hoped that this information will contribute to further research to determine the true economic cost of care and cost of cure of HCV. We hope that our research will aid in strategy and treatment development that will allow future patients the best opportunity to receive efficacious care with optimally controlled side effects while wisely managing our limited healthcare resources.
The authors would like to acknowledge Margie Waits, RN, and Sheree Thomas, BS, for their contributions to this research project.