Diagnostic Potential of Five Different Biomechanical Parameters to Detect Sclerotic Cutaneous Graft-Versus-Host Disease

INTRODUCTION

Skin sclerosis is a significant cause of morbidity in chronic graft-versus-host disease (cGVHD). The current accepted standard for response assessment is the NIH skin score (0-3 scale) which is based on extent of skin involvement and clinician assessment of degree of sclerosis (superficial vs. deep; movable vs. hidebound). There is an urgent need to develop objective measures of skin sclerosis to enable accurate monitoring of disease. The Myoton device is able in a single rapid measurement to simultaneously measure five biomechanical parameters in soft tissue: dynamic stiffness, mechanical stress relaxation time, tissue oscillation frequency, logarithmic decrement, and creep. Sclerotic cGVHD patients have been shown to have increased skin stiffness, relative to normal individuals, in a small study that lacked post-transplant controls.¹ In a cross-sectional study including both cGVHD patients and post-transplant controls, we report the ability to differentiate sclerotic from unaffected post-transplant skin for all five available biomechanical parameters.

METHOD

Sclerotic cGVHD patients (NIH skin score 2-3) (n=13), post-BMT controls (n=10), and healthy controls (n=14) were measured with the Myoton device modified for skin biomechanical assessment. Each subject was measured on 10 bilateral sites (shin, dorsal forearm, volar forearm, upper arm, shoulder, chest, abdomen, upper back, lower back, calf) for a total of 20 measurement sites. For each of the five biomechanical parameters, the overall value per subject was calculated as the average over all 20 measurement sites. We performed receiver operating characteristic (ROC) and area under the curve (AUC) analyses to compare the diagnostic sensitivity of these parameters.

RESULTS

Sclerotic cGVHD patients showed significant increases (p<0.05) in stiffness and frequency and decreases (p<0.05) in relaxation time compared to controls. The ROC and AUC analyses revealed that the overall frequency (sensitivity: 91%; specificity: 93%; AUC: 0.92), stiffness (sensitivity: 91%; specificity: 93%; AUC: 0.92), and relaxation time (sensitivity: 100%; specificity: 82%; AUC: 0.85) allow for high accuracy in the differentiation of sclerotic patients from post-BMT controls. Multivariate analysis suggested that stiffness, frequency, and relaxation time provide more value in combination than any parameters alone in the differentiation of sclerotic patients and post-BMT controls. By contrast, decrement and creep likely do not add significant value beyond the other three parameters. The Pearson correlation for each pair of the 3 key parameters (frequency, stiffness, and relaxation time) in post-BMT patients was lowest (R² = 0.651) for stiffness versus relaxation time, suggesting that this combination provides the highest diagnostic yield. Further, within cGVHD patients, all parameters except decrement were able in univariate analysis to distinguish between a site of affected and unaffected skin on the shin and dorsal forearm (p<0.05).

CONCLUSIONS

Frequency, stiffness and relaxation time of skin demonstrated high accuracy in the differentiation of sclerotic GVHD patients from post-BMT controls. Larger study is needed to develop a multivariate predictive model based on the combination of stiffness and relaxation time. Prospective longitudinal study is also warranted to assess the ability of the Myoton to monitor progression and treatment response in cGVHD patients as compared to the NIH skin score.

ACKNOWLEDGEMENTS

This work is partially supported by Career Development Award Number IK2 CX001785 from the United Sates Department of Veterans Affairs Clinical Science R&D (CSRD) Service.

REFERENCES:

1. Chen F, Dellalana LE, Gandelman JS, et al. Non-invasive measurement of sclerosis in cutaneous cGVHD patients with the handheld device Myoton: a cross-sectional study. Bone marrow transplantation. 2019; 54(4):616