Chronic Graft-Versus-Host Disease: Therapeutics at Last?

Chronic graft-versus-host disease (cGVHD) is a leading cause of late nonrelapse mortality and morbidity following allogeneic hematopoietic stem cell transplantation, and its incidence is rising as recipient age and use of peripheral blood stem cell grafts increase.^1-3  The cardinal feature of cGVHD is fibrosis, which leads to the characteristic clinical manifestations of cutaneous scleroderma and bronchiolitis obliterans syndrome that typically have limited reversibility in their late stages.^4,5  For decades, systemic corticosteroids, with or without calcineurin inhibitors, have been the mainstay of treatment for moderate-to-severe cGVHD, beyond which there have been no established second-line agents available.⁶  Thus, patients usually receive multiple lines of treatment, typically in an ad hoc fashion based on physician experience and biases. This lack of logical therapeutic paradigms for steroid refractory disease has in large part reflected our lack of understanding of the pathophysiology of cGVHD. Fortunately, this has dramatically changed in the past five years^4,7,8  such that we now have a flourishing pipeline of potentially active agents undergoing analysis in both early- and late-phase clinical trials.

The factors initiating acute GVHD and cGVHD are linked: Both have their genesis in the expansion and differentiation of naïve alloreactive T cells along Th1/Tc1, Th17/Tc17, and T follicular helper (T_FH) paradigms, driven by high levels of IL-12 and of IL-6.^9-11  Th17 cells are central to cGVHD pathophysiology: They traffic to GVHD target organs and secrete multiple cytokines including IL-17, IL-21, IL-22, interferon-γ, tumor necrosis factor, colony-stimulating factor (CSF) –1, and GM-CSF stem cell mobilization with G-CSF promotes Th17/Tc17 differentiation and sclerodermatous cGVHD.¹²  Recently, IL-22 has also been implicated in cutaneous cGVHD.¹³  CSF-1 plays a key role in cGVHD pathogenesis by recruiting monocytes and promoting their differentiation into pathogenic M2 macrophages, which mediate fibrosis in lung and skin through secretion of transforming growth factor–β and downstream myofibroblast activation and collagen production.¹⁴  The B-cell pathway is also markedly dysregulated in cGVHD, which is supported by T_FH cells via their secretion of IL-21. This leads to germinal centre B-cell expansion and the secretion of auto- and alloreactive antibodies. Alloantibodies have been shown to be pathogenic in preclinical models of bronchiolitis oblierans¹⁵  and scleroderma.¹⁶  This complex interaction between the key players (Th17/Tc17 and T_FH, germinal center B cells, and tissue macrophages) is modulated by an array of immune regulatory populations. Of these, CD4⁺CD25⁺Foxp3⁺ regulatory T cells (Tregs) are the most extensively studied and are both quantitatively and qualitatively abnormal in cGVHD,¹⁷  but type 1 regulatory cells (Tr1),¹⁸  myeloid-derived suppressor cells (MDSC),¹⁹  and other cell types can also contribute.²⁰ 

The arrival of a range of kinase inhibitors has provided new opportunities for targeting both the innate and adaptive arms of cGVHD pathogenesis. The Bruton tyrosine kinase (BTK) inhibitor ibrutinib was the frontrunner, and in 2017, this agent became the first (and to date, the only) drug to be approved by the U.S. Food and Drug Administration for treatment of cGVHD. Approval was based on the results of a phase Ib/2 study that demonstrated an overall response rate of 67 percent in patients who had failed at least one line of prior therapy.²¹  Although B cells are the primary target of BTK inhibition, ibrutinib also inhibits IL-2 inducible kinase (ITK), which is involved in T-cell activation and myeloid cell recruitment and cytokine response. Additionally, BTK is involved in CSF-1 signaling,²²  and both BTK and ITK inhibition were shown to be important in ibrutinib’s effect on cGVHD.²³  Ibrutinib treatment is associated with a marked reduction in the number of pre–germinal-center B cells and with a fall in IgM level, but with preservation of IgG levels, which is consistent with a lack of plasma cell depletion.²⁴  A phase III study to evaluate the addition of ibrutinib to standard corticosteroids in newly diagnosed moderate-to-severe cGVHD is currently in progress (NCT02959944).

KD025 is an orally available inhibitor of Rho-associated coiled-coil kinase 2 (ROCK2), that decreases the production of IL-17 and IL-21 by T cells.²⁵  It has been shown to reduce the activation of Th17 and T_FH molecules (STAT3, RORγ, and BCL6) and can reverse established cGVHD in murine models.²⁶  In a phase IIa study enrolling patients with cGVHD after one to three prior lines of systemic therapy, the authors reported preliminary response rates of approximately 63 percent and observed a reduction in Th17 numbers (NCT02841995).²⁷  Ruxolitinib is a JAK1/2 inhibitor that had efficacy in steroid-refractory cGVHD in retrospective reports.^28,29  Prospective phase III studies are underway (NCT03112603 and NCT02396628), and a phase I/II study is also ongoing for another JAK1/2 inhibitor, baricitinib (NCT02759731). These agents are associated with myeloid suppression, so there is interest in whether selective JAK1 inhibitors may retain efficacy without invoking myelosuppression, and a phase III study using itacitinib is also in progress (NCT03584516). Spleen tyrosine kinase (SYK) regulates T- and B-cell signaling, and SYK expression is increased on B cells during cGVHD.³⁰  A phase II randomized double-blind study using the SYK inhibitor entospletinib was recently terminated (NCT02701634), and a phase I study using another SYK inhibitor, fostamatinib, is ongoing (NCT02611063).

Proteasome inhibitors are potentially active in cGVHD: In addition to their effect on plasma cells, they can also deplete alloreactive T cells and inhibit NFκB signaling.³¹  A phase II study of bortezomib plus prednisone as upfront therapy for cGVHD showed 80 percent response rate.³²  In phase II studies on refractory cGVHD, carfilzomib did not improve six-month treatment-failure rate (a composite endpoint including death, relapse, and requirement for an additional line of systemic immune suppressive therapy) above historical benchmarks (40% vs. 44%)³³ ; however, a similar study using the orally available proteasome inhibitor ixazomib did show a benefit (28% vs. 44%), with a 34 percent response rate at six months.³⁴  In addition to these B cell– and T cell–directed approaches, another interesting approach has been initiated based on our understanding of the critical role of donor CSF-1R⁺ macrophages in the terminal phase of cGVHD pathophysiology. This is a phase I study evaluating SNDX-6352, a high-affinity antibody blocking CSF-1R in patients who have received at least two lines of prior therapy including ibrutinib (NCT03604692).

Therapeutic approaches aimed at restoring defective immune regulatory cell populations have the theoretical advantage of inducing transplant tolerance and reducing or obviating drug therapy. CD4⁺CD25⁺Foxp3⁺ Tregs are the most well-studied. Low-dose IL-2 administed subcutaneously can expand Treg numbers in vivo and ameliorate cGVHD in 50 to 60 percent of patients³⁵ ; however, this effect is dependent on ongoing treatment, and the response is largely restricted to patients with normal proportions of Tregs in circulation, corresponding to patients who are within one year of cGVHD onset.^35,36  Combining IL-2 with Treg adoptive transfer may overcome this latter difficulty.³⁷  Likewise, new IL-2 muteins are now being studied that have more favorable pharmacokinetic profiles (NCT03422627).

Chronic GVHD has been the scourge of transplantation, but a confluence of factors, including advances in our understanding of cGVHD biology; availability of a range of small molecules, biologics, and cellular therapeutics; and the development of a system for clinical classification, staging, and response assessment,^38,39  has paved the way for a range of highly promising therapeutics that are now on the horizon. The next challenge will be in selecting the right (expensive) drug for an individual at the right time to effectively interrupt the disease process.