Comparative Efficacy Gordon Schiff MD Cook County (Stroger) Hospital with Jay Duhig UIC College of Pharmacy Monday March 12 th 2007.

Comparative Efficacy

Gordon Schiff MD Cook County (Stroger) Hospital

with Jay Duhig UIC College of Pharmacy

Monday March 12th 2007

Ray W and Stein C. N Engl J Med 2006;354:194-201

Principal Limitations of the Present Drug-Regulation ProcessComparative (or relative) Efficacy— a Leading Issue for Drug Policy Reform

Lyrica Meeting w/ Drug Rep

Response

Evidence for Drug Efficacy Requires Comparisons

• Not enough to say “patients responds” to a drug

Evidence for Drug Efficacy Requires Comparisons

• Not enough to say “patients respond” to a drug• 3 concepts, and related trial designs essential for

understanding efficacy

FRAMEWORK- 3 KEY PARADIGMS• Placebo controlled RCT – Does it work?• Comparative efficacy trails – Does it work better

than other altenatives (drug, nondrug)? • N of 1 trials – Does it work in this patient?

Response

Vs. Placebo

Comparison I: Placebo

• Placebo controlled RCT– 1st RCT was MRC Streptomycin 1948

• Need for concurrent randomized controlled groups• Interesting debates over ethics of with-holding streptomycin

to controls– With-holding drug “proven to work” in animals

– Only resolved by historical twist of drug shortage

» Reserved small quantity that was available for TB meningitis and miliary TB

– 1st Origins of placebo controlled trail • Beecher placebo trials

Vs.

I. Other Drugs

II.Same Class Drugs

III.Non Drug Rx

Comparison II: Other Treatments

1. Compare w/ other classes of drugs

2. Compare w/ other drugs in same class

3. Compare w/with non-drug alternatives

vs.

Comparison III: N of 1 Trials

• For later discussion

Copyright ©2003 BMJ Publishing Group Ltd.

MacGillivray, S. et al. BMJ 2003;326:1014

Meta-analyisis: SSRI vs. tricyclic antidepressants

MacGillivray, S. et al. BMJ 2003;326:1014

Comparative Efficacy Azithromycin for Otitis

Comparative Efficacy Azithromycin for Pharyngitis

Comparative Efficacy Azithromycin for Acute Sinusitus

2/2007 Am J Resp Clinical Care Med 

“Comparative efficacy” (and purpose) ……of two types of trials

• Studies asking how good is my drug?– Studies for manufacturer R&D and selling a drug– Starting point is the company well-being– Bias towards showing the value of the drug

• Placebo is usually comparator that shows biggest benefit

– Constitutes majority of drug trials today• Drug industry budget vs. NIH

• Studies asking what is the best therapy?– Fundamentally about comparative efficacy– Starting point is a given problem and patient well-being– Tend to be more complex and harder to perform– Relatively fewer studies of this type– Although what is most needed, growing disproportionately fewer

Venues for Evaluation of Patient Oriented Comparative Efficacy

Existing national effortsNICE --National Institute for Health and Clinical Excellence

CDAC --Canadian Drug Advisory Committee

PBAC - Pharmaceutical Benefits Advisory Committee

Publicly accountableLooking at cost in addition to

efficacy/safetyDrug bulletins, evaluations

Prescrire Cohchrane

Local Formularies

NICE

• UK- National Institute for Health & Clinical Excellence

• Independent organization producing– Public health guidelines– Treatment guidelines– Clinical guidelines

• Published appraisals– http://www.nice.org.uk/guidance/TA/published

2006: w572-w585

Copyright restrictions may apply.

Topol, E. J. JAMA 2004;0:293.3.366-368.

Event Rates of Cardiovascular Death, Myocardial Infarction, and Stroke in the Adenoma Prevention With Celecoxib (APC) Trial

Three myocardial infarctions, all inthe rofecoxib group, were not included in thedata submitted to the Journal. The editors firstbecame aware of the additional myocardial infarctionsin 2001 when updated data were madepublic by the Food and Drug Administration.Until the end of November 2005, we believedthat these were late events that were not knownto the authors in time to be included in the articlepublished in the Journal on November 23, 2000.

NEJM 12/29/2005

It now appears, however, from a memorandumdated July 5, 2000, that was obtained by subpoenain the Vioxx litigation and made available tothe Journal, that at least two of the authors knewabout the three additional myocardial infarctionsat least two weeks before the authors submittedthe first of two revisions and 4 1/2 months beforepublication of the article. Given this memorandum,it appears that there was ample time to includethe data on these three additional infarctionsin the article.

The fact that these three myocardial infarctionswere not included made certain calculationsand conclusions in the article incorrect. Althoughonly summary percentages, not actual numbersof myocardial infarctions, were included in theJournal article, the following tables display thenumerical data without (Table 1) and with (Table2) the three myocardial infarctions.Lack of inclusion of the three events resultedin an understatement of the difference in risk ofmyocardial infarction between the rofecoxib andnaproxen groups (presented in the article as areduction in the risk with naproxen but shownhere as an increase in the risk with rofecoxib).

It contained other data on cardiovascular adverseevents that we believe would have been relevantto the article. We determined from a computerdiskette that some of these data were deleted fromthe VIGOR manuscript two days before it was initiallysubmitted to the Journal on May 18, 2000.Taken together, these inaccuracies and deletionscall into question the integrity of the dataon adverse cardiovascular events in this article.We have asked the authors to submit a correctionto the Journal.

NEJM 12/29/2005

Curfman NEJM 3/16/06

Curfman NEJM 3/16/06

U.S. Deaths from Vioxx More than Vietnam War

• 1/1999--9/2004: 106.7 million rofecoxib prescriptions in US– 17·6% were high-dose, mostly to older patients

• In 2 Merck-sponsored randomised trials: 2,25 relative risks for AMI– 5x for high-dose rofecoxib and 2x for the standard dose – Background rate AMI control NSAID users varied from 7·9 per 1000

person-years in CLASS1 to 12·4 per 1000 person-years in TennCare.

• Using Merck studies relative risks w/ these background rates 88,000– 140,000 excess cases serious coronary disease in US

• Using US national case-fatality rate-44%,suggests thousands of deaths attributable to rofecoxib use (~38,000-61,000)

Graham Lancet 2005

The term “innovation” covers 3 concepts :

- the commercial concept

- the technological concept

- the concept of therapeutic advance

A clear difference between:

• newly marketed substance, or indication, or formulation, etc.

• industrial innovation (chemistry, biotechnology)

• therapeutic advance : « a new treatment that benefits the patient when compared to existing options »– (ISDB Declaration) -

ISDB Declaration on therapeutic advance in the use of medicines (November 2001)

The 3 components of therapeutic advance

1- efficacy

2- safety

3- convenience

Prescrire’s rating system

BRAVOThe drug is a major therapeutic advance in an area where previously no treatment was available

A REAL ADVANCEThe drug is an important therapeutic innovation but has certain limitations

Prescrire’s rating system

OFFERS AN ADVANTAGEThe drug has some value but does not fundamentally change the present therapeutic practice

POSSIBLY HELPFULThe drug has minimal additional value, and should not change prescribing habits except in rare circumstances

Prescrire’s rating system

NOTHING NEWThe drug may be a new substance but is superfluous because it does not add to the clinical possibilities offered by already available treatments

NOT ACCEPTABLEDrug without evident benefit but with potential or real disadvantages

Prescrire’s rating system

JUDGEMENT RESERVEDThe editors postpone their judgement until better data and a more thorough evaluation of the drug are available

23 Years Ratings New Drug “Advances” by Prescrire (1981-2003)

Rating # %Bravo 7 0.2%

A real advance 77 2.7%Offers an advantage 217 7.6%

Possibly helpful 455 15.8%Nothing new 1,913 66.6%

Not acceptable 80 2.8%Judgment reserved 122 4.2%

Total 2,871 100

Comparative Efficacy Does it work compared to nothing

Comparative Efficacy

Does it work: compared to something else

Is it better, how much?

CE is important

• U.S. drug spending is very highXX – Increasing at XX%/year

• How much is spent on new drugs on patent ??– Is this worth it are they better?

– Majority is drug costs– Waste of tax-payer money

• Me-too drugs are the dominant trend– Artificial distinctions– Drug companies like creating a niche (based on packaging dosing)– Be cautious of sub-group analysis and making false

comparison/claims

• CE studies are in the best interest of patients• Other countries have regulatory agencies to provide CE

information

Choices in Exam Room

– Inviduciaula decisionmaking – What should be happening

• Succumbing to marketing the pathof least resistance• Succumbing to the easy solution offered by the drug rep• Weighing pros and cons of the from literature and

assessmewnt of the patients specific risk and benfits• Potential for a real evidence based way to decide if that drug

is benefiting patient and best choice \– N of 1

US FDA

• Very definition of requirements for approval and review – comparative efficacy isn’t on the table – get a qoute from law

• What does IOM say about CE harvest qoutes and references

• Follow up on the article of doctors not following guideilnes

Common Drug Review(CDC)

• Canadian Agency for Drugs and Technology in Health (CADTH)

• Goals:– How does it compare with alternatives?– Which patients will it benefit?– Will it deliver value for money?

Screenshot of clinical guidelines for hypertension - flowchart

Australia’s ExamplePharmacy Benefit Scheme

FDA

• What does the FDA offer to clinicians

• Drug approval letter

• Medwatch safety information

• Ahrq– Brief story re back surgery– Has the potential to fill this role– Mandates of the certs??

Compare sources on LYrica

What does the drug industry think about CE studies?

• Highly subjective

• Advocates just want to spend less

• Requiring CE trials will slow down delivery of new drugs to patients

Sources

• http://www.cadth.ca/media/cdr/complete/cdr_complete_Lyrica_Jan26-06.pdf

• Compare 2 forms of drug information provided by drug regulatory agencies– http://www.ti.ubc.ca/pages/letter61.htm– http://www.fda.gov/cder/foi/appletter/2004/217

23ltr.pdf– NICE?

http://www.mbpct.nhs.uk/medman/prepad/documents/PrescriptionPadJuly2006.pdf

Lancet 1993

Lancet 1993

Prority Medicine (for Europe) WHO 2004 http://mednet3.who.int/prioritymeds/