Background/Purpose: In SSc, treatment responses to tyrosine kinase inhibitors (TKI) are less distinct compared to animal models. Therefore, we assessed whether the heterogeneous therapeutic effects could be related to differences in the activation level of the TKI targets c-abl and PDGFRβ.

Methods: Skin sections of bleomycin treated (n=10), Fra-2 transgenic (tg) (n=6), non-bleomycin treated (n=10) and wt mice (n=6), of diffuse (n=10) and limited (n=28) SSc patients and healthy controls (n=8) were analyzed by Masson’s trichrome stain and immunohistochemistry (IHC) using antibodies against p-PDGFRβ and p-c-abl. Subgroups of bleomycin treated (n=8) and Fra-2 tg mice (n=6) were treated with nilotinib. The bleomycin model reflects early pro-inflammatory stages, whereas the Fra-2 model displays features of vasculopathy-induced fibrosis. Parametric non-related data were expressed as mean±SEM, nonparametric non-related data as median_(Q1,Q3).

Results: In bleomycin treated mice compared to non-bleomycin controls, the dermal expression of both activated PDGFRβ (10.4_(10,11) vs. 3.1_(2,5)% cells/HPF) and c-abl (9.1_(9,10) vs. 1.5_(1,2)% cells/HPF) was strongly upregulated (p<0.05), whereas in Fra-2 tg compared to wt mice, the expression of p-PDGFRβ (8.2_(6,12) vs. 5.4_(4,7)% cells/HPF) and p-c-abl (2.3_(1,6) vs. 5.0_(4,7)% cells/HPF) was not significantly increased (p=0.1). In accordance with the strong activation of PDGFRβ and c-abl in the bleomycin model, nilotinib was effective by reducing the increased dermal thickness (293.5±13.3µm) back (226.4±20.6µm) to non-bleomycin levels (224.3±9.3µm, p<0.05). In Fra-2 tg mice, in accordance with the moderate expression of p-PDGFRβ and p-c-abl, nilotinib did not decrease dermal thickness. Given the correlation of activation level and treatment response in the animal models, we examined potential differences in the dermal activation status in subsets of SSc patients and healthy controls. Compared to healthy controls, in limited SSc, there was no significantly increased expression of p-PDGFRβ (3.1±0.2 vs. 7.0±1.2% cells/HPF, p=0.08) and p-c-abl (1.3±0.3 vs. 1.3±0.3% cells/HPF, p=0.9). Interestingly, in diffuse SSc compared to healthy controls and limited SSc, the activation status of both PDGFRβ (8.0±0.9% cells/HPF) and c-abl (3.9±1.2% cells/HPF) was significantly increased (p<0.05). Whereas in the bleomycin model activated PDGFRβ and c-abl were ubiquitously expressed, particularly in skin fibroblasts, in Fra-2 tg mice there was a vascular predominance which might explain the lacking effect of nilotinib on skin fibrosis. In SSc patients, double staining showed that p-PDGFRβ and p-c-abl were most abundantly expressed in vascular smooth muscle cells (SM22α+) and endothelial cells (vWF+), but only occasionally in scattered dermal fibroblasts(prolylhydroxylase+).

Conclusion: Our study suggests that the activation level and the cellular expression pattern of target molecules are able to predict treatment responses in animal models of fibrosis. The predominant vascular expression of TKI targets in SSc patients might account for the minor anti-fibrotic effects of imatinib in clinical SSc trials.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/differences-in-the-activation-levels-and-expression-patterns-of-the-molecular-targets-of-tyrosine-kinase-inhibitors-may-account-for-the-heterogeneous-treatment-responses/