On 2015 Aug 06, Andrea Messori commented:

Equivalence of sofosbuvir-ledipasvir (12 weeks) and ABT-450/ ritonavir/ ombitasvir/dasabuvir (12 weeks) in previously untreated patients with genotype 1 HCV infection

By Andrea Messori, PharmD, Sabrina Trippoli, PharmD

HTA Unit, Tuscany Region, ESTAR, Regional Health Service, 50100 Firenze, Italy

Some debate is ongoing on which procurement methods can contribute to reduce the high prices of direct-acting antiviral agents (DAAs) indicated for the treatment of hepatitis C (1-4). Two techniques are particularly suitable for this purpose: a) price-volume agreements (3,4); and b) competitive tenders (1,2). In this brief report, we re-examine the results published by Trippoli et al. (5) to evaluate how these findings can be a suitable clinical basis for undertaking competitive tenders according to the current Italian regulation.

At the end of 2012, a national regulation was issued in Italy (“decreto Balduzzi” [6,7]) concerning the acquisition tenders run by our NHS. According to this regulation, when these tenders are aimed at drugs that belong to the same pharmacological class, a preventive authorization must be obtained from our National Medicines Agency (Agenzia del Farmaco, AIFA) to certify that the agents under examination are therapeutically equivalent. Otherwise, these tenders can no longer be performed. Hence, defining equivalence is, in Italy, not only a matter of scientific interest but also a prerequisite for making procurement decisions within our NHS.

In the field of DAAs, the main practical need for conducting tenders is to promote a competition between the two main combination treatments presently available: a) the combination of sofosbuvir and ledipasvir (manufactured by Gilead), abbreviation: SOF/LED; and: b) the four –drug combination manufactured by Abbvie (that includes ABT-450/ ritonavir/ ombitasvir/dasabuvir); abbreviation, ARIOD. To avoid an excessive heterogeneity in the clinical material included in this assessment, we restricted our analysis to the treatments based on a duration of 12 weeks and to the indication of previously untreated patients infected by genotype 1. The end-point was the achievement of sustained virologic response at 12 weeks after the end of therapy (SVR12). It can be seen that these criteria are similar to those employed by Trippoli et al. (5). As regards the margin of equivalence, we adopted the value of ±10% around the overall meta-analytical rate calculated for both treatments.

By running a PubMed query, we searched for more recent studies not included in the meta-analysis of Trippoli et al., but we found none. Furthermore, since the present analysis was essentially based on a proportion meta-analysis design, we replaced the all-in-one Bayesian model adopted by Trippoli et al. with the more traditional random-effect frequentist model of proportion meta-analysis (software: Open Meta-Analyst, version 4.16.12, Tufts University).

Overall, our proportion meta-analysis included 3 studies for SOF-LED and 3 studies for ARIOD (see Appendix 1 for details). Figure 1 shows the Forest plot with the trial-specific rates of SVR12 achievement and the meta-analytic rates estimated separately for SOF-LED and for ARIOD. The overall meta-analytic rate estimated for SOF-LED and ARIOD (6 studies) was 95.5%; hence, the equivalence margin was from 85.5% to 100%. Both SOF/LED (rate of SVR12 achievement, 97.1%; 95% CI of SVR12: 94.4% to 99.8%) and ARIOD (rate of SVR12 achievement, 93.1% ; 95% CI: 85.6% to 100%) satisfied our pre-specified criterion of equivalence. It should be noted that, while most equivalence analyses generally adopt 90%CIs, our analysis employed a more selective confidence interval of 95%.

In conclusion, these results indicate that SOF-LED and ARIOD, given for 12 weeks, are therapeutically equivalent in treatment-naïve patients with genotype 1 infection; this conclusion, formally based on a meta-analysis, confirms the message of equivalence that several experts have reported in narrative terms (1). As regards safety, several recent reviews have examined this point (8-11) and none of them have suggested that there is any difference between these two combination treatments. Hence, this overall picture of the literature indicates that SOF-LED and ARIOD can be included in competitive tenders aimed at the procurement of treatments for patients with the characteristics mentioned above.

References

1. Wilensky G. A New Focus on Prescription Drug Spending. JAMA 2015;314(5):440-441.

2. Brunetto MR, De Luca A, Messori A, Zignego AL. Reducing the price of new hepatitis C drugs in the Tuscany region of Italy. BMJ 2015;350:h3363 doi: 10.1136/bmj.h3363 (Published 24 June 2015), available at http://bmj.com/cgi/content/full/bmj.h3363?ijkey=xYS3zhzXoox8A8t&keytype=ref

3. Messori A. Newest treatments for hepatitis C: how can we manage sustainability? Clin Infect Dis. 2015 Aug pii: civ667. [Epub ahead of print], prepint available at http://www.osservatorioinnovazione.net/papers/cid2015pricing.pdf

4. Messori A. Managing the high cost of innovative drugs: anti-cancer agents vs direct-acting antivirals for hepatitis C (Comment posted 2 April 2015), Ann Intern Med 2015, available at http://annals.org/article.aspx?articleid=2212249#tab

5. Trippoli S, Fadda V, Maratea D, Messori A. Bayesian network meta-analysis to evaluate interferon-free treatments in naïve patients with genotype 1 HCV infection. Eur J Gastroenterol Hepatol 2015 Aug;27(8):983-984

6. Messori A, Fadda V, Maratea D, Gatto R, Trippoli S, De Rosa M, Marinai C. Intravenous proton pump inhibitors for stress ulcer prophylaxis in critically ill patients: determining statistical equivalence according to evidence-based methods. Int J Clin Pharmacol Ther. 2014 Oct;52(10):825-9.

7. Messori A, Fadda V, Gatto R, Maratea D, Trippoli S. Differentiating between “no proof of difference” and “proof of no difference” for new oral anticoagulants. BMJ. 2014; 348: g1955.

8. Kohli A, Shaffer A, Sherman A, Kottilil S. Treatment of hepatitis C: a systematic review. JAMA. 2014 Aug 13;312(6):631-40.

9. Feeney ER, Chung RT. Antiviral treatment of hepatitis C. BMJ. 2014 Jul 348:g3308.

10. Pawlotsky JM. New hepatitis C therapies: the toolbox, strategies, and challenges. Gastroenterology. 2014 May;146(5):1176-92.

11. Ferenci P, Kozbial K, Mandorfer M, Hofer H. HCV targeting of patients with cirrhosis. J Hepatol. 2015 Jun 19. pii: S0168-8278(15)00393-1. doi:10.1016/j.jhep.2015.06.003. [Epub ahead of print]

APPENDIX 1. INFORMATION ON THE TRIALS INCLUDED IN THE PROPORTION META-ANALYSIS.§

AUTHOR SVR12 (n/N) RATE (95% CI) TREATMENT

Kowdley, 206/216, rate=0.954 (0.926 to 0.982), SOF-LED

LONESTAR, 18/19, rate=0.947 (0.847 to 1.048), SOF-LED

Afdal, 211/214, rate=0.986 (0.970 to 1.002), SOF-LED

Kowdley, 70/79, rate=0.886 (0.816 to 0.956), ARIOD

Pearl-IV, 185/205, rate=0.902 (0.862 to 0.943), ARIOD

Pearl-III, 207/209, rate=0.990 (0.977 to 1.004), ARIOD

§The complete bibliographic details of the studies are available from the paper by Trippoli et al. (5). Abbreviations: SOF/LED, sofosbuvir+ledipasvir; ARIOD, ABT-450/ ritonavir/ ombitasvir/dasabuvir; CI, confidence interval.

FIGURE 1. PROPORTION META-ANALYSIS BASED ON A RANDOM-EFFECT MODEL

Rates of SVR12 achievement with sofosbuvir+ledipasvir (12 weeks; 3 studies) and with ABT-450/ ritonavir/ ombitasvir/dasabuvir (12 weeks; 3 studies). The parameter I² is an index of heterogeneity. The Forest plot of this figure is available at http://www.osservatorioinnovazione.net/papers/trippoli-reanalysis.jpg . Abbreviations: SOF/LED, sofosbuvir+ledipasvir; ARIOD, ABT-450/ ritonavir/ ombitasvir/dasabuvir; C.I., confidence interval.

On 2015 Aug 06, Andrea Messori commented:

Equivalence of sofosbuvir-ledipasvir (12 weeks) and ABT-450/ ritonavir/ ombitasvir/dasabuvir (12 weeks) in previously untreated patients with genotype 1 HCV infection

By Andrea Messori, PharmD, Sabrina Trippoli, PharmD

HTA Unit, Tuscany Region, ESTAR, Regional Health Service, 50100 Firenze, Italy

Some debate is ongoing on which procurement methods can contribute to reduce the high prices of direct-acting antiviral agents (DAAs) indicated for the treatment of hepatitis C (1-4). Two techniques are particularly suitable for this purpose: a) price-volume agreements (3,4); and b) competitive tenders (1,2). In this brief report, we re-examine the results published by Trippoli et al. (5) to evaluate how these findings can be a suitable clinical basis for undertaking competitive tenders according to the current Italian regulation.

At the end of 2012, a national regulation was issued in Italy (“decreto Balduzzi” [6,7]) concerning the acquisition tenders run by our NHS. According to this regulation, when these tenders are aimed at drugs that belong to the same pharmacological class, a preventive authorization must be obtained from our National Medicines Agency (Agenzia del Farmaco, AIFA) to certify that the agents under examination are therapeutically equivalent. Otherwise, these tenders can no longer be performed. Hence, defining equivalence is, in Italy, not only a matter of scientific interest but also a prerequisite for making procurement decisions within our NHS.

In the field of DAAs, the main practical need for conducting tenders is to promote a competition between the two main combination treatments presently available: a) the combination of sofosbuvir and ledipasvir (manufactured by Gilead), abbreviation: SOF/LED; and: b) the four –drug combination manufactured by Abbvie (that includes ABT-450/ ritonavir/ ombitasvir/dasabuvir); abbreviation, ARIOD. To avoid an excessive heterogeneity in the clinical material included in this assessment, we restricted our analysis to the treatments based on a duration of 12 weeks and to the indication of previously untreated patients infected by genotype 1. The end-point was the achievement of sustained virologic response at 12 weeks after the end of therapy (SVR12). It can be seen that these criteria are similar to those employed by Trippoli et al. (5). As regards the margin of equivalence, we adopted the value of ±10% around the overall meta-analytical rate calculated for both treatments.

By running a PubMed query, we searched for more recent studies not included in the meta-analysis of Trippoli et al., but we found none. Furthermore, since the present analysis was essentially based on a proportion meta-analysis design, we replaced the all-in-one Bayesian model adopted by Trippoli et al. with the more traditional random-effect frequentist model of proportion meta-analysis (software: Open Meta-Analyst, version 4.16.12, Tufts University).

Overall, our proportion meta-analysis included 3 studies for SOF-LED and 3 studies for ARIOD (see Appendix 1 for details). Figure 1 shows the Forest plot with the trial-specific rates of SVR12 achievement and the meta-analytic rates estimated separately for SOF-LED and for ARIOD. The overall meta-analytic rate estimated for SOF-LED and ARIOD (6 studies) was 95.5%; hence, the equivalence margin was from 85.5% to 100%. Both SOF/LED (rate of SVR12 achievement, 97.1%; 95% CI of SVR12: 94.4% to 99.8%) and ARIOD (rate of SVR12 achievement, 93.1% ; 95% CI: 85.6% to 100%) satisfied our pre-specified criterion of equivalence. It should be noted that, while most equivalence analyses generally adopt 90%CIs, our analysis employed a more selective confidence interval of 95%.

In conclusion, these results indicate that SOF-LED and ARIOD, given for 12 weeks, are therapeutically equivalent in treatment-naïve patients with genotype 1 infection; this conclusion, formally based on a meta-analysis, confirms the message of equivalence that several experts have reported in narrative terms (1). As regards safety, several recent reviews have examined this point (8-11) and none of them have suggested that there is any difference between these two combination treatments. Hence, this overall picture of the literature indicates that SOF-LED and ARIOD can be included in competitive tenders aimed at the procurement of treatments for patients with the characteristics mentioned above.

References

1. Wilensky G. A New Focus on Prescription Drug Spending. JAMA 2015;314(5):440-441.

2. Brunetto MR, De Luca A, Messori A, Zignego AL. Reducing the price of new hepatitis C drugs in the Tuscany region of Italy. BMJ 2015;350:h3363 doi: 10.1136/bmj.h3363 (Published 24 June 2015), available at http://bmj.com/cgi/content/full/bmj.h3363?ijkey=xYS3zhzXoox8A8t&keytype=ref

3. Messori A. Newest treatments for hepatitis C: how can we manage sustainability? Clin Infect Dis. 2015 Aug pii: civ667. [Epub ahead of print], prepint available at http://www.osservatorioinnovazione.net/papers/cid2015pricing.pdf

4. Messori A. Managing the high cost of innovative drugs: anti-cancer agents vs direct-acting antivirals for hepatitis C (Comment posted 2 April 2015), Ann Intern Med 2015, available at http://annals.org/article.aspx?articleid=2212249#tab

5. Trippoli S, Fadda V, Maratea D, Messori A. Bayesian network meta-analysis to evaluate interferon-free treatments in naïve patients with genotype 1 HCV infection. Eur J Gastroenterol Hepatol 2015 Aug;27(8):983-984

6. Messori A, Fadda V, Maratea D, Gatto R, Trippoli S, De Rosa M, Marinai C. Intravenous proton pump inhibitors for stress ulcer prophylaxis in critically ill patients: determining statistical equivalence according to evidence-based methods. Int J Clin Pharmacol Ther. 2014 Oct;52(10):825-9.

7. Messori A, Fadda V, Gatto R, Maratea D, Trippoli S. Differentiating between “no proof of difference” and “proof of no difference” for new oral anticoagulants. BMJ. 2014; 348: g1955.

8. Kohli A, Shaffer A, Sherman A, Kottilil S. Treatment of hepatitis C: a systematic review. JAMA. 2014 Aug 13;312(6):631-40.

9. Feeney ER, Chung RT. Antiviral treatment of hepatitis C. BMJ. 2014 Jul 348:g3308.

10. Pawlotsky JM. New hepatitis C therapies: the toolbox, strategies, and challenges. Gastroenterology. 2014 May;146(5):1176-92.

11. Ferenci P, Kozbial K, Mandorfer M, Hofer H. HCV targeting of patients with cirrhosis. J Hepatol. 2015 Jun 19. pii: S0168-8278(15)00393-1. doi:10.1016/j.jhep.2015.06.003. [Epub ahead of print]

APPENDIX 1. INFORMATION ON THE TRIALS INCLUDED IN THE PROPORTION META-ANALYSIS.§

AUTHOR SVR12 (n/N) RATE (95% CI) TREATMENT

Kowdley, 206/216, rate=0.954 (0.926 to 0.982), SOF-LED

LONESTAR, 18/19, rate=0.947 (0.847 to 1.048), SOF-LED

Afdal, 211/214, rate=0.986 (0.970 to 1.002), SOF-LED

Kowdley, 70/79, rate=0.886 (0.816 to 0.956), ARIOD

Pearl-IV, 185/205, rate=0.902 (0.862 to 0.943), ARIOD

Pearl-III, 207/209, rate=0.990 (0.977 to 1.004), ARIOD

§The complete bibliographic details of the studies are available from the paper by Trippoli et al. (5). Abbreviations: SOF/LED, sofosbuvir+ledipasvir; ARIOD, ABT-450/ ritonavir/ ombitasvir/dasabuvir; CI, confidence interval.

FIGURE 1. PROPORTION META-ANALYSIS BASED ON A RANDOM-EFFECT MODEL

Rates of SVR12 achievement with sofosbuvir+ledipasvir (12 weeks; 3 studies) and with ABT-450/ ritonavir/ ombitasvir/dasabuvir (12 weeks; 3 studies). The parameter I² is an index of heterogeneity. The Forest plot of this figure is available at http://www.osservatorioinnovazione.net/papers/trippoli-reanalysis.jpg . Abbreviations: SOF/LED, sofosbuvir+ledipasvir; ARIOD, ABT-450/ ritonavir/ ombitasvir/dasabuvir; C.I., confidence interval.