1 PIER: 15-Letter Gain From Baseline PIER: Loss of < 15 Letters - - PDF document

1

Neil M. Bressler, MD

Follow-up After Initiating Therapy

• 1. Treat Every Month?
• 2. What is the Efficacy of a Reduced-Frequency

but Fixed Dosing Schedule?

• 3. Other Regimens

Follow-up After Initiating Therapy: PIER: Reduced-Frequency Dosing Schedule of Ranibizumab

0.3 mg (n = 60) 0.5 mg (n = 61) Principal Entry Criteria

• Predominantly classic, minimally classic, or occult with no classic neovascular AMD
• If minimally classic or occult with no classic, evidence of recent disease progression
• VA (Snellen equivalent) 20/40 to 20/320
• N = 184

Sham injection (n = 63)

Primary end point: mean change in VA from baseline to month 12

Ranibizumab

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http://www.healthcareconferencegroup.com/ conferences/63/pier.htm. Accessed July 10, 2006.

Injection frequency: Injections once a month for the first 3 months followed by injections once every 3 months for 24 months

2 Month 8 4 5 6 7 3 9 10 11 12 13 14 15 16 17 18 19 20 21 22 1 23 24

3

4

PIER: Average Loss/Gain of Visual Acuity

• 0.2
• 1.6
• 16.3

Months ETDRS letters

1 2 3 5 6 7 8 9 10 11 12

• 15
• 10
• 5

5 10 16.1-letter difference* 14.8-letter difference†

*P < 0.0001. ETDRS = Early Treatment Diabetic Retinopathy Study.

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http://www.healthcareconferencegroup.com/ conferences/63/pier.htm. Accessed July 10, 2006.

Ranibizumab 0.5 mg (n = 61) Ranibizumab 0.3 mg (n = 60) Sham (n = 63)

4

2

PIER: Loss of < 15 Letters From Baseline at 12 Months

% of patients

Sham injection (n = 63) Ranibizumab 0.3 mg (n = 60) Ranibizumab 0.5 mg (n = 61)

*P < 0.0001 vs sham injection.

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http://www.healthcareconferencegroup.com/ conferences/63/pier.htm. Accessed July 10, 2006.

* *

5

13%

% of subjects

10% 12%*

Sham (n=63) Ranibizumab 0.3 mg (n=60) Ranibizumab 0.5 mg (n=61)

*P=0.87; †P=0.71 vs sham.

PIER: ≥15-Letter Gain From Baseline at Month 12

20 40 60 80 100

†

6

PIER: Adverse Events 12 Month Data

 Adverse ocular events occurring more frequently with ranibizumab vs sham injection not serious

– Conjunctival hemorrhage – Eye pain – Increased intraocular pressure

 No reported cases of endophthalmitis, serious intraocular inflammation, or other ocular serious adverse events  No myocardial infarctions or cerebral vascular events were observed

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http://www.healthcareconferencegroup.com/conferences/63/pier.htm. Accessed July 10, 2006. 7

PIER: Conclusion

 Reduced-frequency dosing schedule reduced risk of vision loss but did not increase chance of vision gain vs sham  Unknown if reduced frequency dosing schedule is inferior to monthly schedule without direct comparison of these treatment regimens

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http://www.healthcareconferencegroup.com/ conferences/63/pier.htm. Accessed July 10, 2006. 8

3

Dosing Intervals When Treating Neovascular AMD:

Monthly? Less Frequently? Dependent on Imaging?

9

PIER: Mean Change in Central Foveal Thickness (μm)

Mean (±SE) change from baseline in Central fovea thickness (μm)

0.5 mg ranibizumab (n=41) 0.3 mg ranibizumab (n=37) Sham Total (n=40) LOCF was used for missing data. No OCT scans in Year 2.

• 175
• 150
• 125
• 100
• 75
• 50
• 25

25 50 75 100 1 2 3 5 8 12

Month

• 126
• 123
• 29

10

4

PIER: Average Loss/Gain of Visual Acuity

• 0.2
• 1.6
• 16.3

Months ETDRS letters

1 2 3 5 6 7 8 9 10 11 12

• 15
• 10
• 5

5 10 16.1-letter difference* 14.8-letter difference †

*P < 0.0001 . ETDRS = Early Treatment Diabetic Retinopathy Study. Brown et al. PIER Study Group. Retina Physician Symposium, 2006 At: http://www.healthcareconferencegroup.com/conferences/63/pier.htm. Accessed July 10, 2006.

Ranibizumab 0.5 mg (n = 61) Ranibizumab 0.3 mg (n = 60) Sham (n = 63)

11

4

PIER: Average Loss/Gain of Visual Acuity

• 0.2
• 1.6
• 16.3

Months ETDRS letters 1 2 3 5 6 7 8 9 10 11 12

• 15
• 10
• 5

5 10 16.1-letter difference* 14.8-letter difference † *P < 0.0001. ETDRS = Early Treatment Diabetic Retinopathy Study.

Brown et al. PIER Study Group. Retina Physician Symposium, 2006. At: http:// www.healthcareconferencegroup.com / conferences/63/pier.htm. Accessed July 10, 2006.

Ranibizumab 0.5 mg (n = 61) Ranibizumab 0.3 mg (n = 60) Sham (n = 63)

Mean Change in Central Foveal Thickness (μm)

Mean (±SE) change from baseline in Central fovea thickness ( ?m)

0.5 mg ranibizumab (n=41) 0.3 mg ranibizumab (n=37) Sham Total (n=40) LOCF was used for missing data. No OCT scans in Year 2.

• 175
• 150
• 125
• 100
• 75
• 50
• 25

25 50 75 100 1 2 3 5 8 12 Month

• 126
• 123
• 29
• 175
• 150
• 125
• 100
• 75
• 50
• 25

25 50 75 100 1 2 3 5 8 12 Month

• 126
• 123
• 29

12

4

Other Questions When Considering Treatment Based on Imaging

 What about results for 40 subjects in PrONTO?

– unknown if results are non-inferior to monthly or PIER regimen, several rules used to determine additional treatment (regular OCTs, periodic FA, VA, combination of attributes)

 What about fluorescein angiography?

– case series show new fluorescein leakage in absence of detection of new thickening or subretinal fluid on OCT

13

Management of Patient and Physician Expectations of AMD Therapy

 Likelihood of at least moderate vision improvement with treatment → not a majority at this time  Likelihood of at least moderate vision improvement without treatment → rare  Risk of moderate or severe vision loss with treatment → rare but not zero  Risk of moderate or severe vision loss without treatment → common

14

Management of Patient and Physician Expectations of AMD Therapy

 Serious injection-related risks → rare, but not zero  Other injection-related risks → common  Rare systemic or other long term risks → unknown  Need for multiple treatments if CNV treatment initiated with unknown date of completion of treatments  Costs vs benefit

15

• 1. Rich et al. Retina. 2006;26:495; 2. Avery et al. Ophthalmology. 2006;113:363; 3. Spaide et al. Retina. 2006;26:383.
• 4. Yoganathon et al Retina;2006:26:994; 5. Shahar et al. Retina. 2006;26:262; 6. Avastin (bevacizumab) PI.

Empiric Use of Bevacizumab in AMD: Safety Considerations

 Safety and efficacy in AMD not evaluated in randomized clinical trial setting, so strength of evidence weaker than ranibizumab1-5  No long-term follow-up has been reported  Systemic side effects in patients with colon cancer include gastrointestinal perforations, hemorrhage, thromboembolic events, hypertension, and proteinuria6 – Unknown relationship to intravitreal injections

16

5

Empiric Use of Bevacizumab in AMD: Safety Considerations

 However, clinical case series suggest that short term effects seem similar to ranibizumab and superior to PDT or pegaptanib  Unknown if bevacizumab is almost as good as (non-inferior to) ranibizumab, somewhat worse, or somewhat better for efficacy or safety  New challenge: should we use ranibizumab or bevacizumab if both are available and no economic difference to patient

17

Does Anyone Know if Bevacizumab is Non-Inferior to Ranibizumab? What is a Non-Inferiority Trial? (EMEA, 27 July 2005)

 Pre-specify/SET a margin of non-inferiority (delta) in the protocol.  What difference in outcome are you willing to give up- and accept as non-inferior?

– The choice of the delta must be justified on clinical and statistical grounds. – It always needs to be tailored specifically to the particular clinical context and no rule can be provided that covers all clinical situations.

18

Does Anyone Know if Bevacizumab is Non-Inferior to Ranibizumab? What is a Non-Inferiority Trial? (EMEA, 27 July 2005)

 Run the study, construct a 2-sided 95% confidence interval for the true difference between the agents.  The interval must lie entirely on the positive side of the non-inferiority margin.

19

Challenges for Non-inferiority Studies

 Testing of the new drug (eg, bevacizumab) would need to mirror the original testing of the Standard drug (eg ranibizumab) – Trial endpoint, design, subject population  Constancy is required to try and ensure new trial outcome of standard treatment is similar to

• utcome in original study

– If ranibizumab outcomes in non-inferiority study are not like ranibizumab outcomes in MARINA and ANCHOR, then you may be sacrificing even more efficacy than you were willing to when you set your margin.

20

6

Can I Tell in My Practice if Bevacizumab is Non-Inferior to Ranibizumab if I Treat 100 New Patients with CNV this Year?

 Pre-specify a margin of non-inferiority

– The choice of the delta must be justified on clinical and statistical grounds and tailored to the clinical context

 Pre-specify Endpoints and Set Margin: For <15

letter loss, anticipate 95% (ranibizumab) and I will accept an outcome of 85% for bevacizumab as non- inferior

 THE MARGIN IS 10%

Translation I am willing to sacrifice a difference of no greater than 10% between the treatment arms to claim bevacizumab is non-inferior to ranibizumab

21

Can I Tell in My Practice if Bevacizumab is Non-Inferior to Ranibizumab if I Treat 100 New Patients with CNV this Year?

 Run the study, construct a 2- sided 95% confidence interval for the true difference between the agents.  The interval must lie entirely

• n the positive side of the pre-

specified margin I ran the study: <15 letter loss is 95% with ranibizumab and 90% with bevacizumab  Difference of -5%  With 50 subjects in each arm, the 95% confidence interval of the difference ranges from +5% to -15% (ie, bevacizumab results are somewhere between 75% and 95% with <15 letter loss); so, FAILS TO MEET non-inferiority

22

New Challenges in the Era of Anti-VEGF Therapy: “PRN”? Ranibizumab or Bevacizumab?

 Determination of optimal treatment schedule unknown

– Monthly vs less frequent dosing schedule – Decisions based on clinical exam vs. OCT images vs. fluorescein angiographic images

 Long-term safety of available anti-VEGF therapies

– Adverse event frequency beyond 2 years will require additional monitoring – Short and long term safety of bevacizumab largely unknown

 CATT may answer many of these questions

23

Potential Benefits of Combining Anti-VEGF Therapy with PDT or Other Rx?

 Better visual acuity outcomes?  Safer (fewer injections)?

– Less ocular risks because fewer intraocular injections – Less systemic risks because fewer intraocular injections – More cost effective because fewer intraocular injections despite increased cost of PDT or other rx

24

7

Evaluate Design of Combination Therapy

25

Primary Endpoint:

Subjects Losing <15 Letters from Baseline at Month 12 % of Subjects

68% 91%* 10 20 30 40 50 60 70 80 90 100 PDT (n=56) PDT + Ranibizumab (n=105) *P =0.0003

FOCUS: PDT compared with PDT plus ranibizumab

13

Primary Endpoint:

Subjects Losing <15 Letters from Baseline at Month 12

*P <0.0001 vs. PDT 64.3 94.3 96.4

10 20 30 40 50 60 70 80 90 100

PDT (n=143) Ranibizumab 0.3 mg (n=140) Ranibizumab 0.5 mg (n=139) %* %*

% of Subjects

%

ANCHOR: PDT compared with ranibizumab without PDT Was it the Addition of Ranibizumab to PDT,

• r Ranibizumab Alone?