SSCQM

Collaborative effectiveness studiesfor rares exposures

Axel FinckhDiv. of Rheumatology & Epidemiology

University of GenevaUniversity of Geneva

Outline1. Rare exposures & cohort studies22. Practical examples of collaborative registry studies

Pan-EU Abatacept analysis Impact of obesity Impact of obesity

2DAG

Schematic diagram of concurrent, retrospective, and ambidirectional cohort studiesSchematic diagram of concurrent, retrospective, and ambidirectional cohort studies

David A Grimes , Kenneth F Schulz. Cohort studies: marching towards outcomes. The Lancet Volume 359, Issue 9303 2002 341 - 345

Cohort Studies - prospectiveAdvantages: Exposure is measured before disease onset (unbiased ) ( ) Rare exposures can be examined Multiple outcomes can be studied for one exposure Multiple outcomes can be studied for one exposure Incidence of disease can be measured (calculate RR) Di d t Disadvantages: Choosing appropriate controls is often difficult Changes over time in exposure status Classification of disease may be influenced by exposure Losses to follow-up (differential) may introduce serious bias

4Elwood M. 3rd Ed Oxford University Press; Oxford: 2007. pp. 1–570

David Neto, Axel Finckh, Florenzo Iannone, Estíbaliz Loza, Elisabeth Lie, Piet L.C. Van Riel, Merete L. Hetland, Karel Pavelka, Jacques-Eric Gottenberg Helena Canhão Xavier Mariette and Carl TuressonGottenberg, Helena Canhão, Xavier Mariette and Carl Turesson

Abstract N° 2910

Supported by an unrestricted research grant by Bristol Supported by an unrestricted research grant by Bristol-Myers SquibbI di id l i ti t Individual investigators:

DN: none AF : Abbvie BMS Pfizer Roche AF : Abbvie, BMS, Pfizer, Roche FI: BMS, Pfizer, Abbvie, UCB, Merck, Roche, Actelion E. Loza: Roche E. Lie: BMS, Pfizer, Abbvie, Roche KP: AbbVie, BMS, MSD, Pfizer, Roche PLC VR: None PLC VR: None MLH: Roche, MSD, Pfizer JEG: Abbvie, BMS, MSD, Pfizer, Roche

XM Pfi R h BMD GSK LFB XM: Pfizer, Roche, BMD, GSK, LFB CT: Abbvie, BMS, Janssen, MSD, Pfizer, Roche, UCB

? Are there differences among pts initiating Abatacept (ABA) across Europe

? What is the impact of these differences on drug effectiveness

A l t ti l h t it i t i iti ti Analyze potential heterogeneity in pts initiating ABA across different European countries and th i t f th h t it ll dthe impact of the heterogeneity on overall drug retention

NORDMARD

9 RA i i

ARTISDANBIO

9 RA european registries: ARTIS, ATTRA, BIOBADASER, DANBIO GISEA NORDMARD ORADANBIO, GISEA, NORDMARD, ORA, SCQM, Reuma.PT

l

ORAReuma.PT

Inclusion criteria were: - Diagnosis of RA

I iti ti f ABA t t t

SCQM

BIOBADASER- Initiation of ABA treatment Primary endpoint: drug retention of ABA / registry

GISEABIOBADASER

ABA drug retention was analyzed using Kaplan-Meier curves and multivariate Cox regressionsgadjusting for potential confounders

Results: Study population

Registers♦ F‐U 

[pt-yrs ]F‐U /pt °

[pt-yrs ]Male 

%Age [Yrs]

RF %

Anti‐CCP

Dis Durat. 

HAQ DAS28 Smoker%

BMI[kg/m2]

CRP [mg/L]

ESR[mm/h]

N° pastc‐DMARDs°

N° pastBiologics°[p y ] [p y ] [ ] CCP

% [Yr][ g ] [ g ] [ ] Biologics

Combined(3834 PTS)

5421 0.9[0.4-2.2]

18.9 57.1±13.1

74.0 67.7 11.3±8.1

1.3±0.6

5.0±1.3

23.1 25.9±5.2

23.5±35.6

33.0±24.8

2 [1-4]

2 [1-3]

NOR‐DMARD (52 PTS) 50 0.6

[0.3-1.0]11.5 51.3

±12.559.6 48.9 14.8

±9.70.9

±0.55.4

±1.123.1 24.0

±4.124.2

±34.236.4

±28.13

[2-4]2

[2-3]SCQM

(506 PTS) 333 0.3[0.1-0.9]

21.7 57.1±13.1

71.6 63.2 9.9 ±8.7

1.1±0.5

4.2±1.0

24 25.9±5.1

13.9±16.3

25.9±20.3

1 [0-2]

1 [0-2][ ] [ ] [ ]

ATTRA(215 PTS) 341 1.2

[0.5-2.6]20.9 50.1

±12.570.1 74.7 11.2

±7.71.5

±0.55.7

±1.122.9 25.5

±4.925.7

±29.338.3

±24.34

[2-5]_

GISEA(375 PTS) 4760 1.0 13.3 56.5 73.6 81.7 10.6 1.4 5.0 22.3 25.8 39.0 34.4 2 1 ( )

[0.4-2.0] ±12.5 ±8.4 ±0.8 ±1.2 ±5.0 ±66.0 ±23.4 [2-3] [1-2]ORA

(1032 PTS) 1750 1.3[0.5-2.8]

20.9 58.1±13.6

71.3 69.8 13.6±9.4

1.2±0.7

5.3±1.2

_ _ 25.0±33.0

35.6±27.7

3[2-4] 2

[1-3]ARTIS

(1019 PTS) 1531 1.0[0.5-2.1]

20.7 58.6±12.4

_ _ 9.6±3.9

1.3 ±0.7

5.1±1.3

57.5 24.8±1.5

19.6±26.5

30.4±23.0

1[1-2]

2 [1-3]

DANBIO(315 PTS) 411 0.7

[0.3-2.0]19 56.0

±12.584.3 59.3 11.4

±9.61.4

±0.74.9

±1.263.5 26.3

±5.618.3

±26.2_ 4

[3-6]1

[1-2][ ] [ ] [ ]BIOBADASER

(283 PTS) * 484 1.4[0.6-2.5]

18.4 56.6±13.1

83.7 60 12.1±8.4

1.7±0.7

5.1±1.6

13.6 _ 18.1±26.0

33.6±27.7

_ 2 [1-3]

Reuma.PT(37PTS) 45 1.1 13.5 59.0 57.1 51.6 12.3 1.6 5.5 23.5 27.2 15.4 40.6 1 2 ( )

[1.4-1.5] ±14.1 ±8.3 ±0.7 ±1.6 ±5.7 ±15.4 ±24.8 [1-3] [1-3]

Legend Table: ♦ Variables are expressed in means and Standard Deviation, if not indicated otherwise. # Modified HAQ (MHAQ) instead of HAQ. ° Medians [Interquartile Ranges] *The 40 patients from LOCALREG (Spain) registry were aggregated within BIOBADASER

Results: Crude drug retention per country

1.0

Survival Estimate / register

1.01.01.01.01.01.01.01.0ARTIS

0.80.80.80.80.80.80.80.80.8

ARTISATTRABIOBADASERDANBIOGISEANORDMARDSORA

0.6Even

t

0.60.60.60.60.60.60.60.6

ORAREUMAPTSCQM

logRank test: P<0.001

0.4eedo

m fr

om

0.40.40.40.40.40.40.40.4

50%

0.2

Fre

0.20.20.20.20.20.20.20.2

0.00.00.00.00.00.00.00.00.0 3819 1827 1057 562 236 66Total Patients at Risk

0 1 2 3 4 5

0.0

Years After first Admin

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

R i t iMST

d dj dRegistriesPts Events

Med [IQR][yrs]

CrudeHR*

ARTIS 1019 4851.9

[1 6 2 2]1.0[ f]

AdjustedHR *°1.0[ f]ARTIS 1019 485 [1.6‐2.2] [ref]

ATTRA 215 753.1

[2.6‐3.9]0.68

[0.54 – 0.85]

2 5 0 87

[ref]

0.89[0.67 – 1.18]

0 81BIOBADASER 283 135

2.5[1.7 – 3.2]

0.87[0.73 – 1.05]

DANBIO 314 1821.2

[1.0 – 1.7]1.39

[1.17 – 1.67]

0.81[0.65 – 1.00]

1.40[1.16 – 1.69]

GISEA 375 90>5

[3.7 – . ]0.56

[0.44 ‐ 0.70]

NORDMARDS0.9 1.89

0.57[0.46 ‐ 0.72]

1.68NORDMARDS 52 33 [0.5 – 2.0] [1.30 – 2.75]

ORA 1018 6281.7

[1.5 – 1.9]1.13

[1.0 1– 1.28]

[1.16 – 2.44]

1.07[0.93 – 1.23]

0 9Reuma.PT 37 19

1.3[1.0 – .]

1.22[0.77 – 1.93]

SCQM 506 2340.9

[0 8 – 1 1]2.01

[1 73 – 2 35]

0.97[0.56 – 1.67]

2.12[1 79 – 2 51]Q 506 234 [0.8 1.1] [1.73 2.35] [1.79 2.51]

* HR > 1 means higher discontinuation °Multivariate adjustments for demographic variables (age, gender), disease characteristics (RF, DAS28 at baseline, Disease duration) and treatment characteristics (N°of prior biologic failures, calendar year of treatment initiation, ABA approval in BIO naïve patients)

Results: Adjusted drug retention per country

1.01.0

Adjusted Survival estimates

1.01.01.01.01.01.01.01.0ARTIS

0.80.80.80.80.80.80.80.80.80.8

ARTISATTRABIOBADASERDANBIOGISEANORDMARDSORA

0.60.60.6Eve

nt

0.60.60.60.60.60.60.6

ORAREUMAPTSCQM

0.40.40.4reed

om fr

om E

0.40.40.40.40.40.40.4

50%

0.20.20.2

Fr

0.20.20.20.20.20.20.2

0.00.00.00.00.00.00.00.00.00.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

Years since the 1st ABA administration

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

Results: Adjusted drug retention per country

1.01.0

Adjusted Survival estimates

1.01.01.01.01.01.01.01.0ARTIS

0.80.80.80.80.80.80.80.80.80.8

ARTISATTRABIOBADASERDANBIOGISEANORDMARDSORA

GDP per capita (€) 2011 rank :

1. Norway (~44000Euros)2. Switzerland (~35000Euros)

0.60.60.6Eve

nt

0.60.60.60.60.60.60.6

ORAREUMAPTSCQM

( )3. Danmark (~32000Euros)4. Sweden (~32000Euros)5. France (~25000Euros)6. Italy(~25000Euros)7 Spain (~24000Euros)

0.40.40.4reed

om fr

om E

0.40.40.40.40.40.40.4

50%

7. Spain (~24000Euros)8. Czech Rep (~20000Euros)9. Portugal (~19000Euros)

0.20.20.2

Fr

0.20.20.20.20.20.20.2

GDP (> 30’000):1.43 (1.29 - 1.58)

0.00.00.00.00.00.00.00.00.00.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

Years since the 1st ABA administration

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

0 1 2 3 4 5

0.0

Patient characteristics at ABA initiation varied Patient characteristics at ABA initiation varied across European countries, probably reflecting differences in eligibility criteria and g yprescription patterns

Large differences in ABA drug retention with a Large differences in ABA drug retention, with a trend to shorter ABA maintenance in countries with relatively liberal access to biologicswith relatively liberal access to biologics

National differences need to be accounted for h l l d d f lwhen analyzing pooled data from several

national registries

• Objective: To analyze the impact of obesity on RA disease• Objective: To analyze the impact of obesity on RA disease activity in patients initiating their 1st biologic agent

• Exposure Variable: WHO BMI categories: BMI ≥ 18.5, < 25: “normal weight” 8 5, 5 o a e g BMI ≥ 25, < 30: “overweight” BMI ≥ 30 < 35: “obese class I” BMI ≥ 30, < 35: obese class I

BMI ≥ 35: “obese class II”

European Congress of Rheumatology EULAR 2013 in Madrid, Abstract FRI0099

Swiss cohort BMI <18-5-25 BMI 25-30 BMI 30-35 BMI >35(SCQM‐RA) (N=1339) (N=818) (N=318) (N=106)

55 58 58 55age 55(43 – 63)

58(49 – 66)

58(49 – 65)

55(49 – 60)

% Female 83 67 70 84% Female 83 67 70 84% RF+ 77 75 64 76% Anti-CCP 66 65 55 60% Steroids 53 55 58 56% Steroids 53 55 58 56% DMARDs 81 81 84 81

US cohort BMI 18 5-25 BMI 25-30 BMI 30-35 BMI≥35US cohort

(CORRONA)BMI 18.5-25

(N=1113)BMI 25-30(N=1334)

BMI 30-35(N=799)

BMI≥35(N=797)

age 57 58 57 55age 57 58 57 55% Female 83 71 73 85% RF 75 77 75 71% RF+ 75 77 75 71% anti-CCP+ 71 71 65 65% Steroid 35 35 33 31% DMARDs 79 80 80 81

1

Association of BMI with DAS28 remission at 12 Mo follow-up

0.8

1

0.6

OR

0.4

Response rates were adjusted for potential confounders using logistic regression. CORRONA and SCQM were analyzed independently with similar models and results compared.

Conclusions

• Data from t o different pop lations indicate that• Data from two different populations indicate that obesity is a risk factor for inferior response to biologic

t d h t d t ti i ti t ithagents and shorter drug retention in patients with longstanding RA

• It is uncertain whether this is explained by suboptimal dosing of these therapies in obese patients or by adosing of these therapies in obese patients or by a true biologic effect of adipose tissue

20

Collaborative (effectiveness) studies for Collaborative (effectiveness) studies for rares exposuresp

• Large collaborative studies are well suited to study rare exposuresg y p

• Large collaborative studies may allow examining multiple potential effects of a single exposure and testing of multiple hypotheses

• BUT, the possibility of bias relating to multiple comparisons means that analyses and results should be hypothesis driven and biologically plausible

21

Thank you !!!Thank you !!!

http://upload.wikimedia.org/wikipedia/commons/e/e6/Views_of_Geneva.jpg

Results: Crude drug retention by stop reasonsKaplan Meier Analysis

1.0

Kaplan Meier Analysis

1.01.01.0OverallIneffectiveness

0.80.80.80.8

IneffectivenessAdverse EventsOther Reason or Remission

0.6m

Eve

nt0.60.60.6

0.4

Free

dom

fro

0.40.40.4

50%

0.2

F

0.20.20.2MST = 1.75yrs [1.61-1.91]

0.00.00.00.0 3819 1827 1057 562 236 66 14Total Patients at Risk

0 1 2 3 4 5 6

Years After first Admin

0 1 2 3 4 5 60 1 2 3 4 5 60 1 2 3 4 5 6

Assumption: Countries with high income => easy access to Bio (+ refunded by health insurance) => frequent switch?

HR robust se lower .95 upper .95 P.VALUE 

AgeBaseline 0.996 0.002 0.993 1.000 0.051g

Gender (F) 1.006 0.061 0.892 1.134 0.922

GDP_group (2)* 1.428 0.051 1.291 1.579 <0.001

PastBio (1) 1.665 0.099 1.370 2.024 <0.001

PastBio (2) 1.672 0.098 1.380 2.027 <0.001

PastBio (3) 1.834 0.101 1.504 2.238 <0.001

PastBio (4) 1.998 0.108 1.615 2.471 <0.001

DAS28 1 055 0 019 1 016 1 096 0 005DAS28 1.055 0.019 1.016 1.096 0.005

DiseaseDurationBaseline 1.001 0.003 0.994 1.007 0.794

RF (1) 0.734 0.063 0.649 0.831 <0.001

ABA_approv 1.599 0.131 1.237 2.067 <0.001

btw0809 1.197 0.060 1.064 1.346 0.003

bef08 1.183 0.081 1.009 1.387 0.039

*GDP group=2 : High income (SCQM, NORDMARDS, DANBIO, ARTIS)_g p g ( Q , , , )ref is low and med incomesHR of GDP_group (2) = 1.4 =>Stop ABA earlier than GDP_group (1)