Discrepancies in hidradenitis suppurativa lesion characterization by providers and patients

BACKGROUND

The hidradenitis suppurativa (HS) clinical response (HiSCR) has come under scrutiny as several HS clinical trials failed to meet primary endpoints with high placebo responses. This may be due to limitations of the tool and raters' ability to accurately characterize and count lesions, rather than lack of efficacy of the studied drug. Due to HS lesion complexity and potential differences in rater training, it was hypothesized that there would be discrepancies in how providers characterize and count lesions for HS clinical trials.

OBJECTIVE

To evaluate how HS providers and patients name and count HS lesions and to identify discrepancies among providers to initiate the development of consensus-driven guidance for HS rater training.

METHODS

An online survey was distributed to the members of HIdradenitis SuppuraTiva cORe outcomes set International Collaboration (HISTORIC). Respondents were asked to classify lesion images composed of multiple and different morphology types and answer questions regarding inclusion of associated dermatological conditions.

RESULTS

Forty-seven HISTORIC members responded (29 providers; 18 patients). There was variability in how respondents classified HS lesions. Of 12 questions containing images, four had ≥50% of respondents choosing the same answer. With an image of a lesion composed of different morphologies, 45% of providers counted it as a single lesion and 45% counted it as multiple distinct lesions. With an image of multiple interconnected draining tunnels, 7% of providers classified it as a single draining tunnel while 79% categorized it as multiple draining tunnels with the number estimated by visual inspection. There was also variability in deciding whether lesions occurring in associated conditions should be considered separately or included in HS lesion counts. Patient responses were also variable.

CONCLUSIONS

The result of the current study reaffirms the gap in how providers characterize and count HS lesions for clinical trials and the need to develop consensus-driven rater training related to HS outcome measures.

Automated classification of hidradenitis suppurativa disease severity by convolutional neural network analyses using calibrated clinical images

BACKGROUND

The assessment of hidradenitis suppurativa (HS) severity requires detailed, and error-prone lesion counts. This proof-of-concept study aimed to automatically classify HS disease severity using machine learning of clinical smartphone images.

METHODS

777 ambient-light and size-controlled images were used to build a class-balanced synthetic dataset (n = 7675). Convolutional neural networks (CNN) were used for automated severity classification (scale 0-3), and to assess disease-dynamics. International Hidradenitis Suppurativa Severity Score System (IHS4) served as reference. A U-NET algorithm was implemented for automated localization of diseased skin.

RESULTS

CNNs were able to distinguish no/mild from moderate/severe disease with an overall prediction accuracy of 78% [receiver operating curve (AUC) 0.85]. Correct IHS4 classification was achieved with an overall accuracy of 72% (AUC 0.84-0.89). In addition, disease dynamics using IHS4 numerical values aligned with CNN outputs (NRMSE 0.262). The UNET algorithm localized lesions with a pixel accuracy of 88.1% and test loss of 0.42.

LIMITATIONS

Limitations in assessing tattooed and hairy skin. Limited number of patients with dark skin colour and Hurley I.

CONCLUSION

CNNs were able to distinguish no/mild from moderate/severe disease, classify disease severity over time, and automatically identify diseased skin areas and the skin phototype. This study breaks new grounds for fast, reliable, reproducible and easy-to-use HS severity assessments using clinical images.

Outcome Measures for the Evaluation of Treatment Response in Hidradenitis Suppurativa for Clinical Practice: A HiSTORIC Consensus Statement

Importance

Although several clinician- and patient-reported outcome measures have been developed for trials in hidradenitis suppurativa (HS), there is currently no consensus on which measures are best suited for use in clinical practice. Identifying validated and feasible measures applicable to the practice setting has the potential to optimize treatment strategies and generate generalizable evidence that may inform treatment guidelines.

Objective

To establish consensus on a core set of clinician- and patient-reported outcome measures recommended for use in clinical practice and to establish the appropriate interval within which these measures should be applied.

Evidence Review

Clinician- and patient-reported HS measures and studies describing their psychometric properties were identified through literature reviews. Identified measures comprised an item reduction survey and subsequent electronic Delphi (e-Delphi) consensus rounds. In each consensus round, a summary of outcome measure components and scoring methods was provided to participants. Experts were provided with feasibility characteristics of clinician measures to aid selection. Consensus was achieved if at least 67% of respondents agreed with use of a measure in clinical practice.

Findings

Among HS experts, response rates for item reduction, e-Delphi round 1, and e-Delphi round 2 surveys were 76.4% (42 of 55), 90.5% (38 of 42), and 92.9% (39 of 42), respectively; among patient research partners (PRPs), response rates were 70.8% (17 of 24), 100% (17 of 17), and 82.4% (14 of 17), respectively. The majority of experts across rounds were practicing dermatologists with 18 to 19 years of clinical experience. In the final e-Delphi round, most PRPs were female (12 [85.7%] vs 2 males [11.8%]) and aged 30 to 49 years. In the final e-Delphi round, HS experts and PRPs agreed with the use of the HS Investigator Global Assessment (28 [71.8%]) and HS Quality of Life score (13 [92.9%]), respectively. The most expert-preferred assessment interval in which to apply these measures was 3 months (27 [69.2%]).

Conclusions and Relevance

An international group of HS experts and PRPs achieved consensus on a core set of HS measures suitable for use in clinical practice. Consistent use of these measures may lead to more accurate assessments of HS disease activity and life outcomes, facilitating shared treatment decision-making in the practice setting.

Assessment tools and phenotype classification for hidradenitis suppurativa

Hidradenitis suppurativa (HS) is a heterogeneous chronic relapsing skin disease. Several assessment tools are used to assess disease severity and to classify disease phenotype; however, no consensus exists. This review evaluates the various assessment tools and phenotypes, assessing their validity and reliability. Numerous assessment tools and phenotype classifications have been proposed for identifying various subtypes within the hidradenitis suppurativa disease spectrum. Each has a different purpose, such as use in daily practice or in clinical trial settings. Several assessment tools and phenotype classifications have been validated but not always with satisfactory results and often with studies showing divergent intra-rater reliability results. A consensus is needed for a validated, easy-to-use, and timesaving assessment tool for routine daily practice. For clinical trials, a validated and extensive assessment tool that also measures response to treatment is also needed.

Overview and comparison of the clinical scores in hidradenitis suppurativa: A real-life clinical data

Introduction

Partly due to its clinical heterogeneity, hidradenitis suppurativa (HS) is difficult to score accurately; illustrated by the large number of disease scores. In 2016, a systematic review by Ingram et al. reported the use of about thirty scores, and since then, this number has increased further. Our aim is twofold: to provide a succinct but detailed narrative review of the scores used to date, and to compare these scores with each other for individual patients.

Materials and methods

The review of the literature was done among articles in English and French, on Google, Google scholar, Pubmed, ScienceDirect and Cochrane. To illustrate the differences between scores, data from some Belgian patients included in the European Registry for HS were selected. A first series of patients compares the severity of the following scores: Hurley, Hurley Staging refined, three versions of Sartorius score (2003, 2007, 2009), Hidradenitis Suppurativa Physician Global Assessment (HS-PGA), International Hidradenitis Suppurativa Severity Scoring System (IHS4), Severity Assessment of Hidradenitis Suppurativa (SAHS), Hidradenitis Suppurativa Severity Index (HSSI), Acne Inversa Severity Index (AISI), the Static Metascore, and one score that is not specific to HS: Dermatology Life Quality Index (DLQI). A second set of patients illustrates how some scores change over time and with treatment: Hurley, Hurley Staging refined, Sartorius 2003, Sartorius 2007, HS-PGA, IHS4, SAHS, AISI, Hidradenitis Suppurativa Clinical Response (HiSCR), the very new iHS4-55, the Dynamic Metascore, and DLQI.

Results

Nineteen scores are detailed in this overview. We illustrate that for some patients, the scores do not predictably and consistently correlate with each other, either in an evaluation of the severity at a time-point t, or in the evaluation of the response to a treatment. Some patients in this cohort may be considered responders according to some scores, but non-responders according to others. The clinical heterogeneity of the disease, represented by its many phenotypes, seems partly to explain this difference.

Conclusion

These examples illustrate how the choice of a score can lead to different interpretations of the response to a treatment, or even potentially change the results of a randomized clinical trial.

Adalimumab efficacy is inversely correlated with body mass index (BMI) in hidradenitis suppurativa

BACKGROUND

Adalimumab is the only FDA-approved biologic for hidradenitis suppurativa (HS). In the setting of increasing obesity rates worldwide, the relationship between adalimumab efficacy for HS and BMI is essential to understand. We assessed this relationship through markers of disease severity and inflammation.

METHODS

Institutional review board-approved retrospective chart review of Montefiore/Einstein HS Center (HSC) patients (n = 57) treated with adalimumab. The relationship between BMI and adalimumab efficacy was assessed through disease severity (HS-Physician Global Assessment [HS-PGA] 0 and Numerical Rating Scale Pain [NRS-Pain]) and inflammatory markers (erythrocyte sedimentation rate [ESR], C-reactive protein [CRP], and interleukin-6 [IL-6]). A BMI ≥ 30 is defined as obese; BMI < 30 is defined as non-obese.

RESULTS

The mean age was 35.8 ± 13.0 years. After adalimumab therapy, those with BMI < 30 experienced significant reductions in HS-PGA (-1.5 ± 0.9; P < 0.0001) and NRS-Pain (-1.6 ± 2.1; P < 0.0001), as well as mean decreases in inflammatory markers ESR, CRP, and IL-6 (-17.90 ± 23.6, -0.71 ± 1.4, -5.88 ± 7.9, respectively; P > 0.05). Obese patients (BMI ≥ 30) experienced mean increases in HS-PGA (+0.22 ± 0.8; P > 0.05) and NRS-Pain scores (+1.41 ± 3.5; P > 0.05) as well as mean increases in ESR, CRP, and IL-6 (+2.62 ± 28.3, +0.44 ± 3.0, +2.35 ± 6.9, respectively; P > 0.05). Comparing the cohorts, differences in changes in HS-PGA, NRS-Pain, ESR, and IL-6 after therapy are significantly different (P < 0.05).

CONCLUSIONS

We identified significantly lower efficacy of adalimumab in HS patients with BMI ≥ 30 compared to those with BMI < 30. Those with BMI ≥ 30 demonstrated signs of both clinical and physiological deterioration while on adalimumab. Future studies are needed to examine adalimumab dosing for HS patients with high BMI, as well as a critical reconsideration of weight-based therapies.

The Prevalence of Periodontitis and Assessment of Oral Micro-Biota in Patients with Hidradenitis Suppurativa: A Descriptive Cross-Sectional Study

Periodontitis has been causally connected with the development of other immune-mediated inflammatory disorders previously. Nevertheless, the current literature does not provide knowledge on oral health in hidradenitis suppurativa (HS) individuals. The aim of this study was to assess the prevalence of periodontitis and characterize an oral microbiome in HS patients. Fifty-five patients with HS and fifty-five healthy controls were enlisted in the study. The incidence of periodontitis was assessed in all patients during the periodontal evaluation. RT-PCR tests were used to quantification of bacterial content and assess the number and composition of nine crucial periodontal pathogens. HS patients had a significantly higher prevalence of periodontitis than healthy controls (45.5% versus 14.5%). Significantly higher values of average copy-count numbers of total bacteria were found in HS patients. The majority of periodontal pathogens were more frequently isolated in patients with HS than among controls. The most frequently detected pathogen in the HS group was Treponema denticola (70.9%), whereas among controls Capnocytophaga gingivalis (34.5%) was the most common isolate. There was no correlation between HS severity and the number of DNA copies of periodontal bacteria. The findings of this research suggest that periodontitis may contribute to the development of HS.

Serum inflammatory markers and leukocyte profiles accurately describe hidradenitis suppurativa disease severity

BACKGROUND

Inflammatory markers and leukocyte profiles have not been longitudinally evaluated as objective signs of hidradenitis suppurativa (HS) severity. We sought to assess C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), interleukin-6 (IL-6), and leukocyte profiles as reliable indicators of HS severity.

METHODS

Retrospective cohort study of 404 patients seen at the Einstein/Montefiore HS Center, Bronx, New York, between March 2019 and November 2020. Associations of disease severity (HS-Physician Global Assessment) with inflammatory markers and leukocyte profiles were assessed by odds ratios (OR) and 95% confidence intervals (95% CI) incorporating up to four visits per patient, adjusting for baseline gender, age, BMI, and smoking status.

RESULTS

Patients with severe disease had elevated CRP (OR 1.87; 95% CI 1.49, 2.34), ESR (OR 1.04; 95% CI 1.03, 1.04), IL-6 (OR 1.08; 95% CI 1.00, 1.16), leukocytes (OR 1.22; 95% CI 1.14, 1.31), neutrophils (OR 1.31; 95% CI 1.20, 1.42), eosinophils (OR 14.40; 95% CI 2.97, 69.74), basophils (OR 2.53; 95% CI 1.09, 5.85), monocytes (OR 5.36; 95% CI 2.49, 11.53), and neutrophil-lymphocyte ratios (OR 1.63; 95% CI 1.35, 1.96) but decreased lymphocytes (OR 0.86; 95% CI 0.68, 1.10).

CONCLUSIONS

This novel longitudinal study of inflammatory markers and leukocyte profiles offers critical laboratory measures to confirm clinically determined disease severity over time.

Hidradenitis Suppurativa Area and Severity Index Revised (HASI-R): psychometric property assessment

BACKGROUND

Validated, reliable, globally accepted outcome measurement instruments for hidradenitis suppurativa (HS) are needed. Current tools to measure the physical signs domain for HS rely on lesion counts, which are time-consuming and unreliable.

OBJECTIVES

To assess the reliability and validity of the Hidradenitis suppurativa Area and Severity Index Revised (HASI-R) tool, a novel method for assessing HS severity, incorporating signs of inflammation and body surface area involved.

METHODS

The measurement properties of the HASI-R tool were evaluated. The tool was created by combining the previously published HASI and Severity and Area Score for Hidradenitis instruments. Twenty raters evaluated 15 patients with HS in a hospital-based ambulatory dermatology clinic. The objectives of the study were to assess inter- and intra-rater reliability of the HASI-R and its components, as well as its construct and known-groups validity. Existing lesion count-based clinician-reported measures of HS and their components were also assessed. Raters were also asked their preferences regarding the various HS severity assessment tools.

RESULTS

The HASI-R had moderate inter-rater reliability [intra-class correlation coefficients (ICC) 0·60]. This was better than all other HS physical sign outcome measures evaluated, which had poor inter-rater reliability (ICC < 0·5). HASI-R had the highest intra-rater reliability (ICC 0·91). The HASI-R had good construct validity and demonstrated known-groups validity. The HASI-R was also the most preferred tool by all raters.

CONCLUSIONS

Results from the clinometric assessment of the HASI-R are encouraging, and support continued evaluation of this clinician-reported outcome measure.