Psoriasis, cardiovascular events, cancer risk and alcohol use: evidence-based recommendations based on systematic review and expert opinion

Authors


  • Conflicts of interest

    • All the authors have been paid consultants of AbbVie. In addition, T. Barnetche has been a speaker for AbbVie, Roche-Chugai. B. Cribier has been paid for consulting activities for Pfizer, Amgen and for writing activities by Leo Pharma and Janssen-Cilag. D. Jullien has been a consultant for Janssen-Cilag, Novartis, Pfizer, MSD. L. Misery has been a paid consultant of Novartis, Janssen-Cilag, Leo Pharma, Pfizer and Pierre Fabre. J.-P. Ortonne has been an investigator, speaker and advisor for Schering-Plough/MSD, AbbVie, Merck Serono, Centocor, Pfizer, Janssen-Cilag, Pierre Fabre, Galderma, Leo Pharma and Meda. C. Paul has been an investigator and consultant for Amgen, Celgene, Janssen-Cilag, Leo, Lilly, Novartis, Pierre Fabre and Pfizer. M.-A. Richard has been an investigator for Amgen, Lilly, Pfizer, Novartis, Janssen, Leo Pharma and a paid consultant for MSD, Pfizer, Novartis, Janssen, AbbVie and Leo Pharma.
  • Funding sources

    • AbbVie provided financial support for publication, but took no further part in the study. The authors have no financial interest in the subject matter or materials discussed in the manuscript.

Correspondence: M.-A. Richard. E-mail: mrichard@ap-hm.fr

Abstract

The relationship between psoriasis, chronic inflammation, cardiovascular risk and risk of cancer has long been debated. In addition, it has been suggested that alcohol consumption may be a risk factor for psoriasis onset and severity.

The aim of this study was to develop evidence-based recommendations on the risk of comorbidities and its management for daily clinical use, focusing on cardiovascular risk, risk of cancer and alcohol use in psoriasis.

A scientific committee identified and selected through the Delphi method clinically relevant questions about cardiovascular risk, risk of cancer and alcohol use in psoriasis. To address these questions, a systematic literature search was performed in Medline, Embase and the Cochrane Library databases. Systematic literature reviews including meta-analysis whenever possible were performed. Subsequently, an Expert board meeting involving 39 dermatologists took place to analyse the evidence and to elaborate recommendations on the selected questions. Recommendations were graded according to the Oxford level of evidence grading system. The degree of agreement of these recommendations was assessed on a 10-point scale, as well as their potential impact on daily clinical practice.

A total of 3242 articles were identified through the systematic literature searches, among which 110 were included in the systematic reviews. Overall, 12 recommendations were elaborated regarding comorbidities management in psoriasis patients. A moderate increased risk of cardiovascular diseases (CVD), mainly myocardial infarction (MI) [meta-analysis of cohort studies: OR = 1.25 (95% CI 1.03–1.52) and of cross-sectional studies: OR = 1.57 (95% CI 1.08–2.27)], and coronary artery disease (CAD) [meta-analysis of cross-sectional: OR = 1.19 (95% CI 1.14–1.24), of cohort studies: OR = 1.20 (95% CI 1.13–1.27) and of case–control studies: OR = 1.84 (95% CI 1.09-3.09)] was acknowledged. This increased cardiovascular risk requires appropriate prevention measures. There was a lack of substantial evidence that conventional systemic treatment has any effect on cardiovascular risk although methotrexate might be cardioprotective. An increased risk of solid cancer potentially associated with smoking and alcohol use was identified. The role of systemic treatment on cancer risk could not be assessed thoroughly due to limited long-term follow-up data. A higher risk of non-melanoma skin cancers especially squamous cell carcinoma was shown, mainly due to previous exposure to oral 8-methoxypsoralen-ultraviolet-A (PUVA), ciclosporin and possibly methotrexate. No firm conclusion could be drawn regarding alcohol and psoriasis due to high variability in alcohol usage assessment in studies. Clinical experience suggests higher alcohol consumption among psoriasis patients compared to the general population. The mean expert participants' level of agreement on these recommendations varied from 6.8 to 9.4.

These 12 recommendations are evidence based and supported by a panel of expert dermatologists. The next step is now to disseminate these recommendations to dermatologists who did not participate in the Expert board meeting and to assess their opinion about the recommendations.

Introduction

Psoriasis is a chronic inflammatory disease of skin, which aetiology involves a complex interaction between genetics and environmental exposures. There is growing emerging evidence that psoriasis patients have a higher prevalence of cardiovascular diseases (CVD), metabolic dysfunction, cancer, chronic inflammatory bowel disease, arthritis, depression or addictions. However, the literature shows conflicting results as some studies identified a strong association between psoriasis and cardiovascular events, mainly myocardial infarction (MI) and stroke and others failed to disclose such an association.

Evidence-based recommendations on cardiovascular risk factors and prevalence of psoriatic arthritis (PsA) in psoriasis were elaborated in 2010 based on systematic review of epidemiological studies.[1-3] The high variability in study setting, data quality regarding cardiovascular risk factors assessment and choice of control group was acknowledged.

As comorbid conditions may impact therapeutic approach in psoriasis, the objective of this study was to elaborate practical recommendations to help practitioners to better appraise the risk of cardiovascular events, cancer and alcohol abuse in psoriasis.

Methods

The elaboration process of these recommendations was divided into two steps. First, clinically relevant questions were built and subjected to systematic literature review (SLR) and extensive evidence-based literature analysis.[4-6] Second, recommendations for daily clinical practice were formulated by a group of dermatologists with a special interest in psoriasis.[1, 7, 8]

In November 2011, the 10 psoriasis experts (CP, SA, FA, BC, PJ, DJ, MLM, LM, MAR, JPO) from the scientific committee selected clinically relevant questions addressing three major domains relevant to psoriasis comorbidities: risk of cardiovascular events, risk of cancer, alcohol consumption in psoriasis. A Delphi voting process was initiated to select the most clinically relevant questions as described before (Table 1). These questions served as a basis for comprehensive SLRs. A bibliographic team including three fellows dermatologists (CH, EB, CP) and their mentors (MAR, LM, CP) were nominated to conduct the SLRs with the help of an experienced biostatistician (TB).

Table 1. Selected questions for each topic
DomainsQuestions
Cardiovascular Morbidity & Mortality(1) Is there an increased risk of cardiovascular morbidity and mortality in psoriasis patients? Can patients subgroups at higher risk be identified to adapt their disease management?
(2) Is psoriasis a cardiovascular morbidity and mortality risk factor independent from traditional risk factors (smoking, hyperlipidaemia, obesity, etc.)? Is it possible to estimate the specific risk associated with psoriasis compared to other cardiovascular risk factors?
(3) Do some psoriasis treatments influence the cardiovascular mortality risk in psoriasis patients?
Alcohol over-consumption(4) Are psoriasis patients known to drink more alcohol than the general population?
Is alcohol a risk factor for psoriasis?
Is alcohol a psoriasis severity factor?
(5) Is alcohol a resistance factor for the treatment of psoriasis?
(6) Does alcohol cessation affect psoriasis? And conversely, is clearance of psoriasis associated with a reduction in alcohol consumption?
Cancer(7) Compared to the general population, is there an increase of the background risk of some cancers in psoriasis patients?
(8) Is the increased risk of cancer associated with the following:
(a) Comorbidities especially smoking and alcohol?
(b) Psoriasis-related chronic inflammation?
(c) Psoriasis treatments?
(9) What are the practical consequences concerning cancer screening and monitoring of psoriasis patients?
(10) Is the cancer-related mortality higher in psoriasis patients than in the general population? Does psoriasis severity have any impact on cancer mortality?

Comprehensive research strategies were developed in collaboration with experienced librarians, using Medline, Embase and the Cochrane Library databases. Relevant key words regarding CVD, alcohol consumption, cancer and psoriasis were used. Publication languages were restricted to English and French, and publication date from 1980 onwards. Additional references were identified via manual searching of reference lists.

Each SLR was conducted in accordance with the updated guidelines of the Cochrane Collaboration[9] by selecting all observational studies prospective and retrospective – including a control group – and evaluating the association between psoriasis and the relevant comorbidities.

The number of articles found and selected for each topic is detailed in Table 2. In each selected article, relevant data were extracted, and when possible meta-analysis was carried out. All computations were performed using Revman 5.1.6 software package developed by the Nordic Cochrane centre (Review Manager (RevMan) [Computer program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011).

Table 2. Results of the systematic literature search for each topic
Recommendation (Number and topic)References retrieved by systematic literature search (n)Articles included in the systematic reviews (n)
1. Cardiovascular morbidity and mortality – Increased risk in psoriasis92933
2. Cardiovascular morbidity and mortality – Risk factor independent of other conventional risk factors00
3. Cardiovascular morbidity and mortality – Psoriasis treatments known to influence2076
4. Are psoriasis patients known to drink more alcohol than the general population?91128
Is alcohol a risk factor for psoriasis?
Is alcohol a psoriasis severity factor?
5. Is alcohol a resistance factor for the treatment of psoriasis?
6. Does alcohol cessation affect psoriasis? And conversely, is clearance of psoriasis associated with a reduction in alcohol consumption?
7. Cancer – Increase of the basic risk of some cancers in psoriasis patients108037
8. Cancer – Risk linked to the following:
 a. Comorbidities especially smoking and alcohol?
 b. Psoriasis-related chronic inflammation?
 c. Psoriasis treatments (biologics not included)?
9. Cancer – What are the practical consequences concerning cancer screening and monitoring of psoriasis patients?
10. Cancer – Is the cancer-related mortality higher in psoriasis patients than in the general population? Does psoriasis severity have any impact on cancer mortality?1156

The results of the SLR and meta-analysis were then summarized for presentation and reviewed by the scientific committee.

In an Expert board meeting held in June 2012, 39 dermatologists with interest in psoriasis reviewed and discussed the evidence available for each question in three workshops, each led by a fellow from the bibliographic team. These discussions aimed at proposing a set of recommendations addressing the clinical questions. At the end of these workshops, the recommendations were presented to the Expert board in a plenary session for agreement and voting.

The 39 participating dermatologists' level of agreement was measured by each recommendation on a 10-point visual analogue scale (1 = no agreement, 10 = full agreement).[10] The potential impact of each recommendation on clinical practice was evaluated by the participants who had to chose between the following statements: ‘this recommendation will change my practice’; ‘this recommendation will not change my practice since it is already in full accordance with my practice’; ‘this recommendation will not change my practice since I do not want to apply it in my practice’.

Subsequently, the evidence grading of each recommendation was assessed according to the Oxford Levels of Evidence. http://www.cebm.net/index.aspx?o=1025

Results

The three SLRs enabled to identify a total of 3242 references, among which 110 relevant articles were included in the analysis. The recommendations are listed in Table 3, with the corresponding evidence grading. The level of agreement of the expert dermatologists regarding the recommendations varied from 6.8 to 9.4.

Table 3. Recommendations
RecommendationsMean agreement (1–10)
Is there an increased risk of cardiovascular morbidity and mortality in psoriasis patients? Can subgroups at higher risk be identified to adapt their disease management? 8.3
 ● There seems to be a slight increase in the risk of cardiovascular morbidity in patients suffering from psoriasis. (Grade B)
 ● There is no substantial evidence for a higher cardiovascular mortality in psoriasis patients. (Grade D)
 ● Young subjects with severe psoriasis may represent a subgroup with a higher risk of myocardial infarction requiring appropriate preventive measures. (Grade B)
Is psoriasis a cardiovascular morbidity and mortality risk factor independent from traditional risk factors (smoking, hyperlipidaemia, obesity, etc.)? Is it possible to estimate the specific risk associated with psoriasis compared to other cardiovascular risk factors? 9.0
 ● There is a lack of robust data to state whether psoriasis is an independent cardiovascular risk factor. (Grade C)
Do some psoriasis treatments influence the cardiovascular mortality risk in psoriasis patients? 6.8
 ● There is no evidence that conventional systemic psoriasis treatments reduce the risk of cardiovascular mortality. (Grade C)
 ● Methotrexate may have a beneficial effect on the risk of cardiovascular diseases; this effect may be enhanced when methotrexate is combined with folic acid. (Grade C)
 ● There is no data indicating a negative effect of etretinate on cardiovascular risk (Grade B). This result can be extrapolated to acitretin (expert consensus). (Grade D)
Are psoriasis patients known to drink more alcohol than the general population? 8.7
 ● Alcohol consumption seems to be higher among psoriasis patients compared to the general population, even though studies are inconsistent. (Grade B)
 ● This increased alcohol consumption does not necessarily meet the criteria for alcoholism. (Grade C)
Is alcohol a risk factor for psoriasis? 7.5
 ● Published data do indicate that increased alcohol consumption is a risk factor for psoriasis onset. (Grade B)
Is alcohol a psoriasis severity factor? 7.7
 ● Published data do not show any correlation between alcohol consumption and psoriasis severity but the methodology used in studies was not appropriate to answer the question. (Grade B)
 ● Clinical experience shows that, in some patients, psoriasis severity is linked to excessive alcohol consumption. (Grade D)
Is alcohol a resistance factor for the treatment of psoriasis? 8.9
 ● There is no published data showing that alcohol is a resistance factor for the treatment of psoriasis. (Grade D)
 ● Clinical experience shows that patients with excessive alcohol consumption are less compliant. (Grade D)
Does alcohol cessation affect psoriasis? And conversely, is clearance of psoriasis associated with a reduction in alcohol consumption? 8.1
 ● There are no published data showing that alcohol cessation influences psoriasis. (Grade D)
 ● Clinical experience suggests that alcohol cessation improves psoriasis. (Grade D)
 ● An improvement in psoriasis may motivate some patients to reduce their alcohol consumption (trust relationship, restoring self-esteem). (Grade D)
Compared to the general population, is there an increase of the background risk of some cancers in psoriasis patients? 8.8
 ● There is a slight increased risk of some cancers: upper aero-digestive tract, liver, lung, pancreatic and urinary tract cancers in psoriasis patients. (Grade B)
 ● The highest increased risk is for skin carcinoma. (Grade B)
 ● There is no increased risk of melanoma. (Grade B)
 ● Regarding lymphoma, misdiagnosis of primary skin lymphoma as psoriasis might have overestimated the risk. (Grade D)
Is the increased risk of cancer associated with: 9.4
(a) Comorbidities especially smoking and alcohol?
(b) Psoriasis-related chronic inflammation?
(c) Psoriasis treatments?
 ● The increased risk of some cancers may be associated with smoking and alcohol consumption. (Grade C)
 ● There is not enough data to evaluate the role of chronic inflammation per se. (Grade D)
 ● The increased risk of skin carcinoma may be associated with PUVA and ciclosporin use. (Grade B)
 ● At doses used in psoriasis, methotrexate is not associated with an increased risk of cancer. (Grade C)
What are the practical consequences concerning cancer screening and monitoring of psoriasis patients? 8.0
 ● Psoriasis does not require a different cancer screening procedure from the general population. (Grade D)
 ● Cancer monitoring and screening must be adapted according to previous and planed psoriasis treatments and also adapted to risk factors for cancer in a given patient: especially alcohol, smoking and sunlight exposure) (Grade D)
Is the cancer-related mortality higher in psoriasis patients than in the general population? Does psoriasis severity have any impact on cancer mortality? 9.0
 ● There is no enough evidence to indicate a higher cancer-related mortality in psoriasis patients. (Grade D)
 ● Psoriasis severity does not seem to be a risk factor for cancer-related mortality. (Grade D)

Cardiovascular morbidity & mortality

The SLR presented in the same issue of this journal focused on the association of psoriasis and PsA with major cardiovascular events, that is, MI, coronary artery disease (CAD) and stroke.[4] Of an initial selection of 929 articles, 33 publications were selected.

Twelve epidemiological studies investigated the association between MI and psoriasis or PsA. A modestly increased prevalence of MI was observed in both psoriasis and PsA patients, which varied with age[11, 12] and psoriasis severity.[11, 13, 14] It remained significantly elevated after controlling for major cardiovascular risk factors. The meta-analysis performed with both cohort and cross-sectional studies showed a higher risk of MI in psoriasis patients compared to the general population: OR = 1.25 (95% CI 1.03–1.52) and OR = 1.57 (95% CI 1.08–2.27) respectively.

Eleven studies showed an elevation of the prevalence of CAD in psoriasis or PsA patients. The meta-analysis disclosed an OR between 1.19 (95% CI 1.14–1.24) for cross-sectional studies and 1.20 (95% CI 1.13–1.27) for cohort studies and 1.84 (95% CI 1.09–3.09) for case–control studies. The risk of CAD seemed to be more pronounced in patients having severe psoriasis and in patients with an early onset of the disease.[15, 17]

To assess the association of psoriasis or PsA with stroke, a total of 13 studies were analysed. Results were inconclusive. The meta-analysis of cross-sectional studies suggested that psoriasis patients were at a slightly higher risk of stroke with an OR of 1.14 (95% CI 1.08–1.99), whereas the meta-analysis of cohort studies failed to show an association with an OR of 1.02 (95% CI 0.92–1.14). Two studies[14, 18] also concluded that patients with psoriasis, particularly if severe, had an increased risk of stroke compared to controls.

There was no robust study to answer the question about the importance of psoriasis as an independent risk factor for cardiovascular morbidity as opposed to the traditional cardiovascular risk factors assessed thoroughly and prospectively.

Overall, the SLR suggested that there was no evidence for a strong increase of cardiovascular mortality in psoriasis and PsA patients. Based on the data review and group discussions, the following recommendations were developed:

Recommendations:

Is there an increased risk of cardiovascular morbidity and mortality in psoriasis patients? Can patients subgroups at higher risk be identified to adapt their disease management?

  • There seems to be a slight increase in the risk of cardiovascular morbidity in patients suffering from psoriasis. (Grade B)
  • There is no substantial evidence for a higher cardiovascular mortality in psoriasis patients. (Grade D)
  • Young subjects with severe psoriasis may represent a subgroup with a higher risk of MI requiring appropriate preventive measures. (Grade B)

Mean experts' level of agreement: 8.3/10

Is psoriasis a cardiovascular morbidity and mortality risk factor independent from traditional risk factors (smoking, hyperlipidaemia, obesity, etc.)?

Is it possible to estimate the specific risk associated with psoriasis compared to other cardiovascular risk factors?

  • There is a lack of robust data to state whether psoriasis is an independent cardiovascular risk factor. (Grade C)

Mean experts' level of agreement: 9.0/10

To assess if psoriasis therapies could play a role in reducing cardiovascular morbidity and mortality in psoriasis, 207 articles were initially identified. Only six articles could be selected. The drugs assessed were methotrexate, etretinate and ciclosporin. None of the observational studies provided with substantial evidence that systemic treatment of psoriasis influences cardiovascular risk although methotrexate may have a favourable effect.

The role of systemic treatments with a potential negative effect on cardiovascular risk (arterial hypertension for ciclosporin or hyperlipidemia with retinoid) and their contribution to cardiovascular morbidity was a matter of discussion within the workshops. Even if the literature review did not show significant cardiovascular risk associated with conventional systemic psoriasis treatments, the retrospective design of the studies and several selection biases prevented from full data interpretation.

For these two questions, the following recommendations were made:

Recommendations:

Do some psoriasis treatments influence the cardiovascular mortality risk in psoriasis patients?

  • There is no evidence that conventional systemic psoriasis treatments reduce the risk of cardiovascular mortality. (Grade C)
  • Methotrexate may have a beneficial effect on the risk of CVD; this effect may be enhanced when methotrexate is combined with folic acid. (Grade C)
  • There is no data indicating a negative effect of etretinate on cardiovascular risk (Grade B). This result can be extrapolated to acitretin (expert consensus). (Grade D)

Mean experts' level of agreement: 6.8/10

Alcohol consumption and psoriasis

The SLR on this topic is presented in the same issue of this journal.[5]

For this research, the initial key words used with ‘Psoriasis’ and ‘Psoriatic arthritis’ were ‘Alcohol related disorders’(Majr) and ‘Alcohol drinking’(Majr) and were then extended to ‘Risk factor’ (Mesh) and ‘Comorbidity’ (Mesh). Over the 911 publications retrieved, 28 were analysed.

Although alcohol consumption has been reported to be linked to psoriasis, the clinical significance remains unclear. Among the 28 selected studies, 18 including four without a control group suggested an association, and five studies did not. The studies did not use the same definition for evaluating alcohol consumption and alcohol abuse, the study design showed high variability. Consequently, a meta-analysis could not be done.

The expert opinion expressed during the working group sessions was in favour of a relationship between alcohol consumption and psoriasis leading to the following recommendations:

Recommendations:

Are psoriasis patients known to drink more alcohol than the general population?

  • Alcohol consumption seems to be higher among psoriasis patients compared to the general population, even though studies are inconsistent. (Grade B)
  • This increased alcohol consumption does not necessarily meet the criteria for alcoholism. (Grade C)

Mean experts' level of agreement: 8.7/10

Alcohol as a risk or severity factor for psoriasis

Limited evidence was found in the literature, mainly because alcohol consumption was not measured before the psoriasis onset and it was not possible to distinguish between cause and consequence. Four studies[19, 22] concluded that alcohol abuse was a risk factor for psoriasis.

The association between alcohol consumption and psoriasis severity was difficult to assess: evaluation of psoriasis severity was not homogeneous in studies. In addition, studies had low power to detect a clinically relevant effect due to small sample size. The following recommendations were formulated:

Recommendations:

Is alcohol a risk factor for psoriasis?

  • Published data do indicate that increased alcohol consumption is a risk factor for psoriasis onset. (Grade B)

Mean experts' level of agreement: 7.5/10

Is alcohol a psoriasis severity factor?

  • Published data do not show any correlation between alcohol consumption and psoriasis severity, but the methodology used in studies was not appropriate to answer the question. (Grade B)
  • Clinical experience shows that, in some patients, psoriasis severity is linked to excessive alcohol consumption. (Grade D)

Mean experts' level of agreement: 7.7/10

Alcohol and resistance to treatment

The experts did not find in the SLR sufficient evidence to provide with clear guidance on the topic. Indeed, two papers[23, 24] investigated alcohol consumption as a resistance factor for psoriasis treatment. However, the small sample sizes prevented from investigating if lower efficacy of medication in alcohol users was due to a direct effect of alcohol or due to the reduction in patient compliance that may be associated with alcohol abuse.

Recommendations:

Is alcohol a resistance factor for the treatment of psoriasis?

  • There is no published data showing that alcohol is a resistance factor for the treatment of psoriasis. (Grade D)
  • Clinical experience shows that patients with excessive alcohol consumption are less compliant. (Grade D)

Mean experts' level of agreement: 8.9/10

Only two case reports[25] reported upon the effect of alcohol cessation on psoriasis. Nevertheless, the experts agreed on the benefits of abstinence from alcohol. In addition, clearance of skin lesions enables to improve body image perception and self-confidence, and could perhaps enable to decrease alcohol consumption. The following recommendations were formulated:

Recommendations:

Does alcohol cessation affect psoriasis?

And conversely, is clearance of psoriasis associated with a reduction in alcohol consumption?

  • There are no published data showing that alcohol cessation influences psoriasis. (Grade D)
  • Clinical experience suggests that alcohol cessation improves psoriasis. (Grade D)
  • An improvement in psoriasis may motivate some patients to reduce their alcohol consumption (trust relationship, restoring self-esteem). (Grade D)

Mean experts' level of agreement: 8.1/10

Cancers

The SLR on this topic is presented in the same issue of this journal.[6] Out of an initial selection of 1080 articles, 37 studies were selected and reviewed.

For all of them, the frequency of cancer in psoriasis patients was compared to the frequency in general population. The analysis was carried out according to the type of cancer and its risk factors. Whenever possible, a meta-analysis was performed.

The overall risk of non-skin cancer appears to be increased in psoriasis in comparison with the general population: standardized incidence ratio (SIR) = 1.16 (95% CI 1.07–1.25). For cancers potentially associated with tobacco and alcohol, the estimated risk is higher in psoriasis than in the general population with the following SIR = 1.52 (95% CI 1.35–1.71) for bronchopulmonary cancers, SIR = 3.05 (95% CI 1.74–5.32) for upper aero-digestive tract cancers, SIR = 1.31 (95% CI 1.11–1.55) for urinary tract cancers, SIR = 1.90 (95% CI 1.48–2.44) for liver cancers and SIR = 1.46 (95% CI 1.10–1.95) for pancreatic cancer. No significant increased in risk was found for colorectal, neurological, or cervix cancers. Only two studies[26, 27] of 12 found a significant higher risk of breast cancer.

Among the 12 studies analysing risk of lymphoma in psoriasis compared to controls, seven showed an association between total lymphoma,[30, 38] Hodgkin lymphoma,[30, 31] non-Hodgkin lymphoma[31, 34] and psoriasis. Two studies[30, 34] specifically assessed the risk of skin lymphoma in psoriasis and found an increased risk of cutaneous T-cell lymphomas (CTCL) associated with psoriasis. However, a misclassification bias between skin lymphoma and psoriasis may have artificially increased the global risk estimation. The risk of leukaemia was not increased.

Concerning skin cancer, both meta-analysis for squamous cell carcinoma and basal cell carcinoma confirmed that psoriasis patients had an increased risk of skin carcinoma compared to the general population, with an overall SIR of 5.31 (95% CI 2.63–10.71) and 2.00 (95% CI 1.83–2.20), respectively, whereas the risk of melanoma was not increased. The following recommendations were formulated:

Recommendations:

Compared to the general population, is there an increase of the background risk of some cancers in psoriasis patients?

  • There is a slight increased risk of some cancers: upper aero-digestive tract, liver, lung, pancreatic and urinary tract cancers in psoriasis patients. (Grade B)
  • The highest increased risk is for skin carcinoma. (Grade B)
  • There is no increased risk of melanoma. (Grade B)
  • Regarding lymphoma, misdiagnosis of primary skin lymphoma as psoriasis might have overestimated the risk. (Grade D)

Mean experts' level of agreement: 8.8/10

Recommendations:

Is the increased risk of cancer associated with the following:

  • (a)Comorbidities especially smoking and alcohol?
  • (b)Psoriasis-related chronic inflammation?
  • (c)Psoriasis treatments?
  • The increased risk of some cancers may be associated with smoking and alcohol consumption. (Grade C)
  • There is not enough data to evaluate the role of chronic inflammation per se. (Grade D)
  • The increased risk of skin carcinoma maybe associated with PUVA and ciclosporin use. (Grade B)
  • At doses used in psoriasis, methotrexate is not associated with an increased risk of cancer. (Grade C)

Mean experts' level of agreement: 9.4/10

To determine the patient cancer screening and surveillance measures, the current European guidelines on systemic treatment of psoriasis[35, 38] were scrutinized. The recommendations were not entirely consistent. The need for skin cancer screening before initiating phototherapy or ciclosporin is recommended. The limitation of the number of phototherapy sessions is required and the association between 8-methoxypsoralen-ultraviolet-A (PUVA) therapy and ciclosporin must be avoided. The experts concluded that the higher risk of cancer in psoriasis may not require any specific screening measure, but prevention measures and monitoring must be adapted according to the therapy chosen. These specific recommendations are in line with those previously validated by the same psoriasis experts' meeting for phototherapy:[8, 39]

Recommendations:

What are the practical consequences concerning cancer screening and monitoring of psoriasis patients?

  • Psoriasis does not require a different cancer screening procedure from the general population. (Grade D)
  • Cancer monitoring and screening must be adapted according to previous and planed psoriasis treatments and also adapted to risk factors for cancer in a given patient: especially alcohol, smoking and sunlight exposure (Grade D)

Mean experts' level of agreement: 8.0/10

Only six of 115 studies addressed the question about cancer mortality in psoriasis. Three studies[40, 42] showed an increased cancer-related mortality in psoriasis inpatients[41, 42] or in patients with severe psoriasis.[40] These populations are highly selected and it is difficult to extrapolate the results to the entire psoriasis patient population. There is no robust evidence to state that cancer mortality is higher in psoriasis patients than in the general population. The following recommendations were formulated:

Recommendations:

Is the cancer-related mortality higher in psoriasis patients than in the general population? Does psoriasis severity have any impact on cancer mortality?

  • There is no enough evidence to indicate a higher cancer-related mortality in psoriasis patients. (Grade D)
  • Psoriasis severity does not seem to be a risk factor for cancer-related mortality. (Grade D)
  • Mean experts' level of agreement: 9.0/10

Discussion

In the present recommendations supported by SLRs, meta-analysis and opinions from psoriasis experts, guidance is provided regarding CVD morbidity and mortality, alcohol and risk of cancer in psoriasis patients.

These 12 recommendations, comprising 29 statements, were graded by psoriasis experts for their degree of evidence, agreement between experts and for their ability to change clinical practice.

It must be acknowledged that SLR and meta-analysis did not provide with high quality evidence to support the recommendations. A large proportion of the formulated statements (14 out of 29) were only supported by Grade D evidence, and nine of these 14 were solely established according to the experts' opinions.

The major limitation of this study is the heterogeneity of the methodological quality of the selected publications: various studies designs, lack of large scale prospective studies, lack of long-term follow-up and the limited information about potential confounding factors. Additional studies are required to determine if the excess in mortality in patients with severe psoriasis is related top psoriasis, its treatment or to comorbidities.

Thus, in this expert meeting, personal clinical practice and experience of experts had to be confronted with SLR, leading to some recommendations based on expert clinical experience rather than literature evidence for three questions. However, the experts' level of agreement concerning all these recommendations was high, with nine recommendations obtaining a score between 8 and 10/10 on individual rating.

During the meeting, experts were not familiar and at ease with cancer screening and management of cardiovascular risk in psoriasis. On the contrary, concerning alcohol consumption expert discussions were very lively and the literature could be complemented by personal experience.

The ability of these recommendations to change the participating experts everyday clinical practice is difficult to measure prospectively. Indeed, building of an evidence-based strategy requires large population-based studies to establish the link between comorbidities and psoriasis. Based upon this study, comorbidities appear to be important to consider and the role of dermatologists in patient evaluation is advocated to assess addictive behaviour, alcohol consumption, cardiovascular and cancer risk in patients.

With these recommendations, dermatologists could be advised to alert psoriasis patients, particularly if their disease is severe, to be screened for cardiovascular and cerebrovascular risk factors, and to be regularly monitored for cancer risk and for psoriasis-associated cardiovascular comorbidities.

Conclusion

Although studies identified increased risk of comorbidities in psoriasis patients, the methodological strength of the existing literature remains low. There is a need to better assess, through large prospective cohort studies with adequate control groups, the relationship between psoriasis and comorbidities.

Acknowledgements

We thank all the participants of the National Meeting*, the librarians who helped carry out the systematic literature searches, and AbbVie for their financial support.

*Marina Alexandre-Audaire (Bobigny); Hélène Aubert-Wastiaux (Nantes); Edouard Begon (Paris); Clara Bioul (Les-Pennes-Mirabeau); Danielle Bouilly-Auvray (Dijon); Thierry Boye (Toulon); Pierre Bravard (Montivilliers); Didier Brive (Tours); Guillaume Chaby (Salouel); Cécile Clément (Caen); Anne-Marie Coutellier (Torcy); Alexiane Dallot (Aulnay-sous-Bois); Michel Dandurand (Nîmes); Valérie Florin (Lille); Florence Fourquet (Allauch); Yvon Gall (Toulouse); Céline Girard (Montpellier); Sophie Grande (Saint-Genis-Laval); Caroline Jacobzone-Lévêque (Lorient); Daniel Jacquemier (Bourg-en-Bresse); Juliette Jegou (Reims); Mathilde Kemula Potel (Créteil); Brigitte Lagrange (Avignon); Morad Lahfa (Toulouse); Renaud Laurans (Marseille); Cristina Livideanu (Toulouse); Dominique Lons-Danic (Paris); Isabelle Lota-Zunino (Aix-en-Provence); François Maccari (Saint-Mandé); Martine Mauboussin-Gautron (Joué-lès-Tours); Catherine Michel (Mulhouse); Jacques Molho (Melun); Josiane Parier (Paris); Marc Perrussel (Rennes); Eve Puzenat (Besançon); Nathalie Quiles-Tsimaratos (Marseille); Pascal Schmoor (Villenave d'Ornon); Pascale Sotias Girault (Clamart); Yolande Veran (Metz).

Ancillary