Co-morbidity and skin disease: The psoriasis model

1
Co-morbidity and skin disease The psoriasis
model

• Joel M. Gelfand, MD, MSCE
• Medical Director, Clinical Studies Unit
• Assistant Professor of Dermatology
• Associate Scholar, Center for Clinical
  Epidemiology
• and Biostatistics
• University of Pennsylvania
• Joel.Gelfand_at_uphs.upenn.edu

2
(No Transcript)
3
Fitzpatricks Dermatology in General Medicine

• Fitz wanted to show the dermatology was relevant
  to medical practice
• Key aspect of this was internal diseases which
  presented in the skin
• Can skin diseases present internally?

4
(No Transcript)
5
(No Transcript)
6
Historical Overview of Psoriasis
Time Period Psoriasis Theory
BC Disease often confused with leprosy and described in Old Testament as a condition rendering one unclean. Entity recognized in writings of Hippocrates (400 BC).
1841 Named psoriasis by von Hebra
1963 Van Scott demonstrates epidermal hyperproliferation
1980s Immune system emerges as important to pathogenesis of psoriasis
2000s Psoriasis is a systemic inflammatory disease?
7
Why might psoriasis be a systemic inflammatory
disease?

1. Immune abnormalities are profound
2. Psoriasis severity is associated with greater
   levels of systemic inflammation (e.g. CRP, Th-1
   cytokines)
3. Inflammation may be a common pathway to a variety
   of diseases including atherosclerosis, obesity,
   and insulin resistance

Extensive psoriasis is characterized by an
estimated 20 billion T cells infiltrating the skin
Krueger JG, Bowcock, A. Ann Rheum Dis
20056430-36.
8
Natural history of psoriasis

• Disease severity
• 85 Mild, 10 Moderate, 5 severe
• Control of severe disease
• 50 of patients intensively treated continue to
  have very active disease (PUVA cohort)
• 75 of patients with severe disease are not
  receiving appropriate therapies (NPF survey)
• Pathways affected and possible outcomes
• Inflammatory atherosclerosis, thrombosis
• Angiogenesis EPC endothelial dysfunction
• Metabolic oxidative stress

Nijsten, T et al, Clinical severity of psoriasis
in last 20 years of PUVA study. Arch Dermatol,
20071431113-21. Gelfand, J.M., Long-term
treatment for severe psoriasis we're halfway
there, with a long way to go. Arch Dermatol,
20071431191-3. Horn, E.J., et al., Are
patients with psoriasis undertreated? Results of
National Psoriasis Foundation survey. J Am Acad
Dermatol, 2007 57957-62. Azfar, R.S. and J.M.
Gelfand, Psoriasis and metabolic disease
epidemiology and pathophysiology. Curr Opin
Rheumatol, 2008.20416-22.
9
Paradigm of the Natural History of Psoriasis and
Co-morbidities

• Outcomes
• Cancer
• Cardiovascular disease
• Metabolic disease
• Arthritis
• Mortality

Risk factors Genes Environment

• Mediating Factors
• Pathophysiology (inflammation,hyper-proliferation
  , angiogenesis)
• Treatment
• Psychosocial impact

10
Why is this important?

• Severe psoriasis
• 50 increased risk of mortality
• 3-5 years of life lost
• Excess mortality not explained by predictors
  routinely measured in medical care

11
New Standard of Care

• At the very least, dermatologists, who may be
  the only health care provider for psoriasis
  patients, must alert these patients to the
  potentially negative effects of their disease as
  it relates to other aspects of their health.
• National Psoriasis Foundation clinical
  consensus on psoriasis co-morbidities and
  recommendations for screening

JAAD 2008581031-42
12
Cardiovascular Disease and Psoriasis

• Psoriasis associated with excess risk of CVD
  since 1960s
• Th-1 inflammation is central to pathogenesis of
  atherosclerosis and MI

Hansson GK NEJM 20053521685-1695
13
Psoriasis is Associated with Cardiovascular risk
factors

• Smoking
• Obesity
• Dyslipidemia
• Hypertension
• Diabetes

Neiman AN, Porter S, and Gelfand JM. Expert
Review of Derm. 2006163-75
14
Prevalence of cardiovascular risk factors
OR (severe vs. control) 1.9 1.3 1.3 1.3 1.3 1.8
Neimann et al. JAAD 200655829-835
15
Diabetes is independently associated with
psoriasis
Model Mild Psoriasis (95 CI) Severe Psoriasis (95 CI)
DiabetesAdjusted for age gender 1.3 1.9
Diabetes adjusted for age, gender hypertension obesity hyperlipidema smoking 1.1 1.6
Incident diabetes in severe psoriasis adjusted
OR 2.6 (95 CI 1.1-5.9) Defined as gt2 years
duration and gt 2 oral psoriasis Rxs per year
Neimann et al. JAAD 200655829-835 Brauchli, YB.
BJD 2008 In press
16
Psoriasis a risk factor for CAD and MI?
Psoriasis
CAD MI
Smoking HTN DM Obesity Lipids
17
Key Question

• Is the association between psoriasis and MI
• Indirect (confounding)
• Bias (study design)
• Direct
• If the association is not explained by
  confounding or bias then psoriasis is a
  RISK FACTOR for MI
• Risk factors may be in the causal pathway of an
  association

18
JAMA. 2006 Oct 11296(14)1735-41
19
Study Design data source

• General practice research database (GPRD) is a
  medical records database established in UK in
  1987
• Data is recorded by GP on diagnoses and
  medications
• Diagnoses and treatments by specialists well
  captured
• Over 9 million patients and gt 40 million person
  years of follow-up data from 1987-2002
• Use of GPRD has been validated for numerous
  medical conditions (psoriasis, MI, smoking, and
  other co-variables)

20
Validation of Exposure/Outcome

• Psoriasis
• Epidemiology and treatment patterns very similar
  to UK estimates
• 90 of patients with a psoriasis code were
  confirmed to have psoriasis 3-4 years later based
  on questionnaire sent to GP (N100)
• MI
• 90 of patients with a code for MI were confirmed
  to have MI based on having 2 of the following
  criteria characteristic chest pain,
  characteristic changes in the electrocardiogram,
  characteristic serial rises in the concentrations
  of cardiac enzymes, an arteriogram documenting a
  recent coronary occlusion, or fibrinolytic
  therapy (N400)

Neimann A et al. JAAD 200655829-835 Meir CR et
al. Lancet 19983511467-71
21
Study design

• Study Design Cohort study
• Age 20-90
• Exposure
• Psoriasis
• Mild
• Severe psoriasis (received systemic therapy)
• Control no history of psoriasis code matched
  from same practice and start date
• Start date max psoriasis, registration
• End Date min endpoint, death, transfer

22
Conceptual Underpinning of Case-control, Cohort,
and Clinical Trials
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Exposed
o
o
allocation process
o
o
o
End of observation period
o
o
o
o
Unexposed
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Study population
study time
? Study outcome
23
Characteristics of Study Population
Variable Control Mild Psoriasis Severe Psoriasis
N () 556995 (80.96) 127139 (18.48) 3837 (0.56)
Gender
Male 261023 (46.86) 61100 (48.06) 1869 (48.71)
Female 295972 (53.14) 66039 (51.85) 1968 (51.29)
P lt 0.001 P 0.011
Age
Mean (median, 25th, 75th Percentile) 45.72 (42, 30, 60) 46.35 (44, 31, 60) 49.75 (49, 36, 63)
P lt 0.001 P lt 0.001
24
Systemic therapies received by patients with
severe psoriasis
Note Percentages do not add to 100 because
patients could have received more then one
systemic therapy.
25
Risk of MI in psoriasis
Control Mild psoriasis Severe psoriasis
MI incidence per 1000 person yrs 3.58 4.04 5.13
Unadjusted relative risk of MI 1.11 (1.07-1.17) 1.43 (1.18-1.72)
26
Adjusted relative risk of MI in psoriasis
patients based on patient age
Age Mild Psoriasis Severe Psoriasis
40 1.2 2.4
60 1.1 1.4
27
Excess risk of MI due to psoriasis
Age Mild Psoriasis Severe Psoriasis
40-50 1 MI per 3646 patients/yr 1 MI per 623 patients/yr
50-60 1 MI per 2147 patients/yr 1 MI per 430 patients/yr
28
Stroke risk in psoriasis patients
Analysis Mild Psoriasis Severe Psoriasis
Primary analysis adjusted for age and gender 1.07 (1.02, 1.12) 1.44 (1.10, 1.88)
Primary analysis adjusted for stroke risk factors 1.06 (1.01, 1.11) 1.43 (1.10, 1.87)
Excess risk of stroke 1 stroke per 4115 mild psoriasis patients 1 stroke per 530 severe psoriasis patients
Gelfand JM et al. J Investigative Dermatol.
2008128S81
29
Sensitivity Analyses

• Information bias
• Patients seen at least once per year
• End of observation was last visit to GP
• Directionality of association
• Excluded patients with h/o MI or stroke
• Excluded events occurring within the first 6
  months
• Disease/Treatment effects
• Exclusion of methotrexate treated patients
• Exclusion of cyclosporine and retinoid treated
  patients
• Exclusion of PsA

30
Limitations

• Unknown or unmeasured confounding variables
• Mild group is heterogeneous
• Skin severity not measured directly
• Use of methotrexate in severe group may
  underestimate the relative risk of MI
• Mechanism not investigated

31
Psoriasis An Independent Risk Factor for
Atherosclerosis
Ludwig RJ Br J Dermatol 2007156271-6
32
Psoriasis and Atherosclerosis Study Design
Methods

• Design Cross-sectional study
• 32 psoriasis patients treated in an inpatient
  setting 32 matched controls
• Prior history of CVD was exclusionary
• Non contrast-enhanced 16-row spiral CT performed
  and Agatston score was calculated

33
Prevalence and Severity of CAD in Psoriasis

• Prevalence of CAD greater in psoriasis patients
• 59 vs. 28, P0.02
• Severity of CAD associated with psoriasis
• Mean CAC 78 in psoriasis vs. 22 in controls,
  (P0.02)
• Severity of psoriasis (based on of
  treatments/yr) correlated with CAC score (r0.29)

34
Psoriasis An Independent Risk Factor for
Atherosclerosis

• Psoriasis independently predicts atherosclerosis
• Controlled for age, sex, hypertension, lipids, FH
  of cardiovascular disease, diabetes, smoking,
  BMI, and CRP
• Psoriasis explained an estimated 8 of the
  variance

35
Limitations

• Small study in highly selected psoriasis
  inpatients vs. controls who were outpatients
• Could not determine impact of disease vs. impact
  of therapy
• Mechanism not investigated

36
Late Breaking News

• Psoriasis is independently associated with
  carotid atherosclerotic disease and impaired
  endothelial function
• Balci DD et al, Increased carotid artery
  intima-media thickness and impaired endothelial
  function in psoriasis JEADV ISSN 1468-3083
• In patients with PsA, psoriasis severity is an
  independent predictor of cardiovascular disease
• Gladman, DD et al. Cardiovascular morbidity in
  psoriatic arthritis. Ann Rheum Dis 2008.094839

37
Conclusion

• Evolving literature identifying
• Burden of cardiovascular risk in patients with
  psoriasis
• Independent effect of psoriasis on DM and CV risk
• More research needed to determine how psoriasis
  severity and activity and psoriasis treatment
  alters the risk of cardiovascular events
• Important implications for the management of
  patients with psoriasis

38
New Questions and Directions

• Which other skin diseases confer excess CV risk?
• What additional approaches can be used to confirm
  existing data?
• What is the role of "unconventional" CV risk
  factors in explaining the risk of CV disease in
  patients with psoriasis such as depression,
  stress, physical inactivity, etc
• What is the magnitude of CV risk in order to
  inform treatment decisions such as ATPIII
• What is the risk of CVD in patients with severe
  disease who are not being treated

39
New Questions and Directions

• Can the risk of CVD attributable to psoriasis be
  modified with treatment - observational vs.
  experimental approaches
• What CV pathways are affected by psoriasis
  activity - endothelial dysfunction? endothelial
  precursor cells? cardiac load? metabolic demand
  and oxidative stress, etc
• What surrogates of cv risk would be most useful
  to study?
• What existing US data sources could be used to
  investigate these associations - who could fund
  these studies?
• How can existing post marketing studies be
  combined to address these questions?

40
Acknowledgements

• Funding through NIH/NIAMS K23 Career development
  support from Dermatology Foundation, the Herzog
  Foundation and the American Skin Association