Study of Oral Treatments for Hepatitis C

Efficacy of Hepatitis C Virus (HCV) Treatment with Zepatier (Elbasvir/Grazobevir) with Ribavirin (RBV) for 16 weeks when used in Genotype 1a patients with Baseline RASs (NS5a Resistance Associated Substitutions or RAPs -Resistance Associated Polymorphisms). Efficacy defined as HCV RNA undetectable 12 weeks post treatment. Table excludes Genotype 1b patients.

Time frame: 12 weeks post treatment

Population: All Genotype 1a patients who started treatment by arm randomized (Elbasvir/Grazobevir) with Baseline RAS at any location (28, 30, 31, or 93)

Fatigue severity collected from validated, Patient Reported Outcomes survey (PROs), 'PROMIS Fatigue Short Form'. PROMIS® T-scores were computated to compare difference between baseline value of PROMIS score to the highest (worst) score during treatment. Results presented as computated t-score from baseline to average of on Treatment Scores. A positive (+) score suggests improvements in functional well-being. A negative (-) PRO change score is suggestive of symptom improvement or lack of drug side effect. PROMIS Fatigue Score scale per question: 1=Never, 2=Rarely, 3=Sometimes, 4=Often, 5=Always with 7 questions for a total maximum score of 35.

Time frame: Baseline and Average On-Treatment Score

Population: Patients with Baseline PROMIS Fatigue score who started treatment by arm as treated (Period 2)

Fatigue severity collected from validated, Patient Reported Outcomes survey (PROs), 'PROMIS Fatigue Short Form'. PROMIS® T-scores were computated to compare difference between baseline value of PROMIS score to the highest (worst) score during treatment. Results presented as computated t-score from baseline to average of on Treatment Scores. A positive (+) score suggests improvements in functional well-being. A negative (-) PRO change score is suggestive of symptom improvement or lack of drug side effect. PROMIS Fatigue Score scale per question: 1=Never, 2=Rarely, 3=Sometimes, 4=Often, 5=Always with 7 questions for a total maximum score of 35.

Time frame: Baseline to On-treatment

Population: Analysis limited to patients who completed baseline and at least one on-treatment Fatigue PRO survey during phase 1 as treated.

HCV-PRO, a survey designed to assess functional status of patients with HCV and measures physical, emotional, and social functioning, productivity, intimacy, and perception of quality of life, was used to assess functional well-being. In general, lower score is worst outcome and higher scores indicate greater well-being and functioning. However, for ease of interpretation, HCV-PRO scale has been transformed by using '100 - HCV-PRO' ultimately revising the score to mean 0 (lowest score) is best to 100 (worst outcome). A positive change (End of treatment to baseline) suggests improvements in functional well-being. Total score = (SUM-N)/(4\*N)\*100, where N is the number of questions answered. End of Treatment -Baseline were used to calculate CHANGE in outcome. Number of subjects reflects participants from both Phase 1 and 2.

Time frame: End of Treatment - Baseline

Population: Patients with completed Baseline and End of Treatment HCV-PRO survey who started treatment by arm as treated (Period 2)

HCV-PRO, a survey for patients with HCV that measures physical, emotional, and social functioning, productivity, intimacy, and perception of quality of life, was used to assess 'overall functioning and well-being'. In general, lower score is worst outcome and higher scores indicate greater well-being and functioning. However, for ease of interpretation, HCV-PRO scale has been transformed by using '100 - HCV-PRO' ultimately revising the score to mean 0 (lowest score) is best to 100 (worst outcome). A positive change (End of treatment to baseline) suggests improvements in functional well-being. Total score = (SUM-N)/(4\*N)\*100, where N is the number of questions answered.

Time frame: Baseline to End of Treatment

Population: Analysis limited to patients who completed baseline and at least one on treatment Fatigue Short form. Evaluation is based on actual HCV regimen received.

Headache was evaluated by the HIT-6 score, a validated, Patient Reported Outcomes survey (PROs) 'PROMIS Headache Impact Test (HIT)' with scores ranging from 36 to 78 with higher score reflecting greater impact. Mean change in headache side effect was evaluated using difference between baseline value of HIT-6 score to the highest (worst) score during treatment. Estimates of mean change and differences obtained from a constrained longitudinal linear mixed-effects model that treated baseline score as one of outcomes. Negative values for mean change represent improvement, while negative values for 'difference' indicate LDV/SOF performed better than PrOD

Time frame: Baseline to On-Treatment

Population: Analysis limited to participants who completed both baseline and end of treatment PRO

Headache was evaluated by the HIT-6 score, a validated, Patient Reported Outcomes survey (PROs) 'PROMIS Headache Impact Test (HIT)' with scores ranging from 36 to 78 with higher score reflecting greater impact. Mean change in headache side effect was evaluated using difference between baseline value of HIT-6 score to the highest (worst) score during treatment. Estimates of mean change and differences obtained from a constrained longitudinal linear mixed-effects model that treated baseline score as one of outcomes. Negative values for mean change represent improvement in symptom.

Time frame: Baseline to On-Treatment

Population: Analysis limited to patients who completed a baseline and on-treatment PRO and as treated (Period 2) during Phase 1 only.

Participants completed the Patient Reported Outcomes surveys (PROs) 'PROMIS Nausea/Vomiting -4 Short Form' at baseline and during treatment. Raw scores are converted to standardized T-scores with a range of 45.0-80.1. Higher scores indicate worse nausea/vomiting. Results presented as mean difference from baseline to average of on Treatment Scores (highest/worst) score during treatment. A negative (-) PROMIS change score is suggestive of symptom improvement or lack of drug side effect. Estimates of mean change were obtained from a constrained longitudinal linear mixed-effects model that treated the baseline score as one of the outcomes.

Time frame: Baseline and Average On-Treatment Score

Population: Patients with Baseline PROMIS Nausea/vomiting score who started treatment by arm as treated and completed baseline and on-treatment survey.

Patients completed the PROMIS® Nausea Short Form at Baseline (T1) and on-treatment. PROMIS raw scores from each of the completed questionnaires were converted to standardized T-scores. Change was calculated as the difference between baseline and on-treatment score. T-scores for the PROMIS Nausea and Vomiting 4a scale range from 45.0 - 80.1. Higher scores indicate worse nausea. Negative values for mean change represent improvement. The estimates of mean change and differences were obtained from a constrained longitudinal linear mixed-effects model that treated the baseline score as one of the outcomes. The model expressed mean score as a function of DAA regimen, cirrhosis status, HCV genotype, sex, age, race, and previous treatment status.

Time frame: Baseline to On-Treatment

Population: Analysis is limited to patients who completed at least one baseline and one end of treatment PRO survey during Phase 1 (end date corresponds to last PrOD patient start date)

Fatigue severity collected from validated, Patient Reported Outcomes survey (PROs), 'PROMIS Fatigue Short Form'. PROMIS® T-scores were computated to compare difference between baseline value of PROMIS score to the highest (worst) score during treatment. Results presented as computated t-score from baseline to average of on Treatment Scores. A positive (+) score suggests improvements in functional well-being. A negative (-) PRO change score is suggestive of symptom improvement or lack of drug side effect. PROMIS Fatigue Score scale per question: 1=Never, 2=Rarely, 3=Sometimes, 4=Often, 5=Always with 7 questions for a total maximum score of 35.

Time frame: Baseline to End of Treatment

Population: Analysis is limited to patients who completed baseline and on-treatment fatigue PRO during Phase 1 (last PrOD patient started). Safety profiles of the evaluated regimens in PRIORITIZE are by actual treatment received.

Fatigue severity collected from validated, Patient Reported Outcomes survey (PROs), 'PROMIS Fatigue Short Form'. PROMIS® T-scores were computated to compare difference between baseline value of PROMIS score to the highest (worst) score during treatment. Results presented as computated t-score from baseline to average of on Treatment Scores. A positive (+) score suggests improvements in functional well-being. A negative (-) PRO change score is suggestive of symptom improvement or lack of drug side effect. PROMIS Fatigue Score scale per question: 1=Never, 2=Rarely, 3=Sometimes, 4=Often, 5=Always with 7 questions for a total maximum score of 35.

Time frame: Baseline-On Treatment (up to 16 weeks)

Population: Analysis limited to Period 2 -as treated, participants who completed baseline and on-treatment surveys.

HCV-PRO, a survey for patients with HCV that measures physical, emotional, and social functioning, productivity, intimacy, and perception of quality of life, was used to assess 'Overall Functioning and Well-being'. In general, lower score is worst outcome and higher scores indicate greater well-being and functioning. However, for ease of interpretation, HCV-PRO scale has been transformed by using '100 - HCV-PRO' ultimately revising the score to mean 0 (lowest score) is best to 100 (worst outcome). A positive change (End of treatment to baseline) suggests improvements in functional well-being. Total score = (SUM-N)/(4\*N)\*100, where N is the number of questions answered.

Time frame: Baseline to End of Treatment

Population: Analysis limited to patients who completed baseline and at least one survey while on treatment up to end of treatment. Evaluation was based on actual regimen received versus regimen to which patient was randomized.

Headache was evaluated by the HIT-6 score, a validated, Patient Reported Outcomes survey (PROs) 'PROMIS Headache Impact Test (HIT)' with scores ranging from 36 to 78 with higher score reflecting greater impact. Median change in headache side effect was evaluated using difference between baseline value of HIT-6 score to the highest (worst) score during treatment. Estimates of mean change and differences obtained from a constrained longitudinal linear mixed-effects model that treated baseline score as one of outcomes. Negative values for change represent improvement, while negative values for 'difference' indicate LDV/SOF performed better than PrOD

Time frame: 12 weeks (Baseline and Average On-treatment Score)

Population: Analysis limited to patients randomized up to last patient randomized to PrOD and participants who completed baseline and on-treatment survey.

Headache was evaluated by the HIT-6 score, a validated, Patient Reported Outcomes survey (PROs) 'PROMIS Headache Impact Test (HIT)' with scores ranging from 36 to 78 with higher score reflecting greater impact. Median change in headache side effect was evaluated using difference between baseline value of HIT-6 score to the highest (worst) score during treatment. Estimates of median change and differences obtained from a constrained longitudinal linear mixed-effects model that treated baseline score as one of outcomes. Negative values for mean change represent improvement, while negative values for 'difference' indicate LDV/SOF performed better than PrOD

Time frame: Baseline -on Treatment (12-16 weeks)

Population: Analysis limited to patients randomized to EBR/GZR regimen or SOF/LDV regimen and participants who completed both baseline and on treatment survey.

Patients completed the PROMIS® Nausea Short Form at Baseline (T1) and on-treatment. PROMIS raw scores from each of the completed questionnaires were converted to standardized T-scores. Change was calculated as the difference between baseline and on-treatment score. T-scores for the PROMIS Nausea and Vomiting 4a scale range from 45.0 - 80.1. Higher scores indicate worse nausea. Negative values for change represent improvement. The estimates of change and differences were obtained from a constrained longitudinal linear mixed-effects model that treated the baseline score as one of the outcomes.

Time frame: Baseline- On Treatment (up to 16 weeks)

Patients completed the PROMIS® Nausea Short Form at Baseline (T1) and on-treatment. PROMIS raw scores from each of the completed questionnaires were converted to standardized T-scores. Change was calculated as the difference between baseline and on-treatment score. T-scores for the PROMIS Nausea and Vomiting 4a scale range from 45.0 - 80.1. Higher scores indicate worse nausea. Negative values for mean change represent improvement. The estimates of change and differences were obtained from a constrained longitudinal linear mixed-effects model that treated the baseline score as one of the outcomes.

Time frame: Baseline to end of treatment

Population: Analysis limited to patients who completed baseline and on treatment Nausea short form during Phase 1 (up to last PrOD randomized) as treated.

SVR (Sustained Virologic Response) 12 will be defined as undetectable hepatitis C virus (HCV) RNA at 12 week follow-up visit (12 -24 weeks after HCV treatment discontinuation as dictated by standard of care at each individual site). Number of subjects reflects participants who started EBR/GZR or SOF/LDV- based treatment (with or without RBV) during Phase 1 and 2.

Time frame: 12-24 weeks post HCV treatment

Population: All mITT without imputation (missing data excluded). Includes only number of participants for whom SVR12 data is available and based on study drug as randomized (Period 1). Participants for whom SVR 12 is not available are excluded from this table.

SVR (Sustained Virologic Response) 12 will be defined as undetectable hepatitis C virus (HCV) RNA at 12 week follow-up visit (12 -24 weeks after HCV treatment discontinuation as dictated by standard of care at each individual site). Number of subjects reflects participants randomized during Phase 1 only.

Time frame: 12 -24 weeks post-treatment

Population: Participants analyzed based on HCV treatment as assigned by randomization.

SVR (Sustained Virologic Response) 12 will be defined as patients who have undetectable hepatitis C virus (HCV) RNA at 12 week follow-up visit (12 -24 weeks after HCV treatment discontinuation as dictated by standard of care at each individual site). mITT with imputation (missing=failure). Total number of subjects reflects participants from Phase 1 only.

Time frame: 12 weeks post-treatment

Population: mITT with imputation, Missing SVR data =failure. Numbers represent participants according to arm randomized (period 1). Population excludes participants who did not start any HCV treatment.

SVR (Sustained Virologic Response) 12 will be defined as undetectable hepatitis C virus (HCV) RNA at 12 week follow-up visit (12 -24 weeks after HCV treatment discontinuation at discretion of provider). mITT with imputation (missing=failure). Total number of subjects reflects participants from EBR/GZR with or without RBV and SOF/LDV with or without RBV randomized during Phase 1 and Phase 2.

Time frame: 12 weeks post-treatment

Population: mITT with imputation, Missing SVR data = Failure. Numbers represent participants according to arm randomized (Period 1). Population excludes participants who did not start any HCV treatment.

HCV-PRO score, a validated PROMIS survey used to evaluate overall functioning and well-being in HCV patients, was utilized to compare long-term 'within treatment' changes of functional well-being. In general, lower score is worst outcome and higher scores indicate greater well-being and functioning. However, for ease of interpretation, HCV-PRO scale has been transformed by using '100 - HCV-PRO' ultimately revising the score to mean 0 (lowest score) is best to 100 (worst outcome). A positive estimate (Post treatment to baseline) suggests baseline functional well-being improvement. Total score = (SUM-N)/(4\*N)\*100, where N is the number of questions answered.

Time frame: Treatment start date up to 2 years post-treatment

Population: Analysis limited to patients who completed HCV-PRO assessments during post-treatment. For this safety analysis, regimen is based on treatment received versus treatment to which patient was randomized. Arms with and without RBV are combined in consideration that usage of RBV was decided by HCV provider and not part of study randomization.

Change in HCV-associated symptoms was calculated as the mean differences of mean scores from multiple surveys from the NIH Patient-Reported Outcomes Measurement Information System (PROMIS)-- Fatigue, nausea, belly pain, sleep disturbance, and diarrhea) and functional status (well-being) when comparing baseline to early post-treatment and late post treatment surveys. Mean change scores were calculated by comparing baseline to early post-treatment (1 yr) and late post-treatment (approximately 3 years) surveys. T-scores for the PROMIS Nausea and Vomiting 4a scale range from 45.0 - 80.1. Higher scores indicate worse nausea. Negative values for mean change represent improvement.Negative numbers suggest better symptoms (improvement in HCV-associated symptoms). PROMIS Fatigue Score scale per question: 1=Never, 2=Rarely, 3=Sometimes, 4=Often, 5=Always with 7 questions for a total maximum score of 35.HCV-PRO positive estimates suggest baseline functional well-being improvement.

Time frame: 1 year post treatment discontinuation (Early post-tx)

Population: Analysis includes all patients who started treatment on either EBR/GZR or SOF/LDV regimen and is limited to patients who completed surveys. Analysis does not decipher between RBV usage given RBV usage was based on HCV provider selection and not study randomization

Number/Percentage of patients with persistence of viral cure, SVR (SVR = Sustained Virologic Response)- defined as undetectable HCV RNA at least 24 weeks following HCV Treatment.

Time frame: 24 weeks post-end of treatment up to 153 weeks

Population: Analysis limited to patients who have HCV RNA result at least 24 weeks after end of treatment regimen as per period 1 (As randomized).

Percentage of Cirrhotic patients with persistence of viral cure, SVR, (SVR= Sustained Virologic Response) defined as undetectable HCV RNA at least 24 weeks following HCV Treatment.

Time frame: Up to 132 weeks post HCV treatment

Population: Analysis limited to patients who had HCV viral load result at least 24 weeks after treatment and analyzed as per randomized (Period 1)

Number of participants who achieved SVR (sustained virologic response), defined as undetectable HCV RNA 12 weeks post-treatment with RASs (Resistant Associated Substitutions) after treatment with EBR/GZR or SOF/LDV regimen

Time frame: 12 weeks post HCV treatment

Population: Analysis is based on as assigned by randomization and patients with available RAS data. Exploratory RAS analysis is grouped by DAA regimen with or without RBV given 1)RBV usage was decided by HCV provider and not part of study randomization and small sample size (limited number of RAS in regimens with RBV)

The number of participants with adverse events that led to early treatment discontinuation (defined as duration less than originally prescribed treatment regimen)

Time frame: Treatment start date through treatment completion (up to 24 weeks)

Population: As randomized and treated population during Phase 1 and 2 (See Period 1). Analysis based on HCV DAA regimen randomization per study. Differentiation between RBV usage was not included in analysis given that RBV was not part of study randomization but rather decided by HCV provider.

Mean change (delta) in FIB-4, an indirect non-invasive measure of liver fibrosis, calculated as baseline median -long term follow up median, following SVR after any of the study treatment regimens. Change in FIB-4 where negative values indicate improvement in liver fibrosis score and positive values indicate worsening of fibrosis score. There is no upper or lower limit for change. FIB-4 = age (years) \* AST(IU/L)/Platelets (10\^3/L) \* ALT\^.5(IU/L). In general, Score of 0-1.29 is low risk for advanced fibrosis, 1.30-1.67: intermediate risk for advanced liver fibrosis, \>2.67: high risk for advanced fibrosis.

Time frame: Baseline to up to 3 years post treatment discontinuation

Population: Population analyzed limited to cirrhotic patients who had two separate FIB-4 values collected as part of standard care and achieved SVR (defined as undetectable HCV 24 weeks post treatment). Given pragmatic trial design, availability of fibrosis markers is limited thereby restricting analysis (substantially limited sample size) to difference in fibrosis markers following SVR with any HCV treatment used in study.

The Voils Medication Adherence Survey (VMAS) was used to evaluate medication adherence during HCV treatment. Participants responded to three questions about the extent of adherence during the past seven days of treatment (during early and late on-treatment occasions). Participants responded using a five-point ordinal scale of missed dosing from 1 (none of the time) to 5 (all of the time). On each occasion participants were coded as being Non-adherent if any response was \> 1, otherwise they were coded as Adherent. Probability estimates of percentage of patients reporting non-adherence were calculated per HCV treatment (Direct Acting Antiviral-DAA) regimen: 1)EBR/GZV (elbasvir/grazoprevir, 2)SOF/LDV (sofosbuvir/ledipasvir), 3)PrOD

Time frame: 12-16 weeks of HCV treatment

Population: Analysis is limited to participants to patients who started treatment prior to January 2017 (last day patient started on PrOD treatment)