The CD4 T helper cells can be classified in to Th1 and Th2 based on cytokine secretion profiling, with Th1 cells secreting IFN-γ, IL-2 and TNFα. Th2 cells produce IL-4,IL-5,IL-6,IL-10 and IL-13. Th1 cells are responsible for delayed type hypersensitivity and are important in defense from infectious microorganisms. When exposed to foreign antigens, the APCs (macrophages and dendritic cells [DCs]) migrate to lymphoid organs and present antigens to naïve T cells and produce proinflammatory cytokines, such as IL-12, which leads to Th1 type response. Human and animal studies have shown that acute GVHD is a clinical syndrome caused by an imbalance between Th1 (pro-inflammatory cytokines) and Th2 response (anti-inflammatory cytokines) [18–20]. Chronic GVHD is a T cell mediated allo-reactive process and Th2 response dominates in cGVHD [21, 22]. Another subset of CD4 T cells which produce IL-17 (Th17) have been also implicated in development of cGVHD. Dander et al. [13] studied the role of IL-17 producing CD4+ T cells (Th17) in cGVHD in serum of 51 patients post-allo HSCT with clinical manifestations of cGVHD and compared this to 15 healthy donors (HD). Patients with cGVHD showed an increase of Th17 population compared with HD (mean SFU = 178/25,000 cells, n = 18, ANOVA P < 0.001). Importantly, by analyzing the proportion of Th17 cells according to the activation status of cGVHD (active vs. inactive phases), the authors were able to demonstrate that patients with active cGVHD show an increase of Th17 population (mean SFU = 237/25,000 cells, n = 13, ANOVA P < 0.001). Inflammatory cytokines produced by Th17 cells such as IL-6, TNF- α and IL-8 were also significantly elevated in patients with active cGVHD.

Therapeutic approaches: Based on the critical role played by CD4+ T cells in propagating cGVHD, attempts have been made to reduce these alloreactive T cells by inhibition of the Janus Kinase/Signal Transducer and Activator of Transcription (JAK/STAT) pathway. Common gamma chain signaling via JAK pathway leads to up-regulation of IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21—this cytokine storm causes T cell activation, lineage commitment, and survival signaling. Patients with steroid refractory cGVHD who were treated with the commercially available JAK1/2 inhibitor ruxolitinib exhibited excellent clinical responses with an overall response rate of 85.4 %. These responses were independent of organ involvement [23]. Additional clinical trials are in progress to further study the role of this important class of medications in prevention of cGVHD (NCT02806375, NCT01790295).

Regulatory T cells, or Tregs, constitute 5–10 % of circulating CD4+ T cells, and suppress auto- and allo-reactive T cell clones. Tregs have also been associated with clinical symptoms of cGVHD. Immunophenotypically, Tregs are CD4+ and CD25+ and express forkhead transcription factor FOXP3. Zorn et al. [10] evaluated CD4+CD25+ Tregs in 30 patients with cGVHD after allogeneic HSCT, 27 patients without active cGVHD, and 26 healthy controls. They also evaluated T cell receptor excision circles (TREC) by peripheral blood polymerase chain reaction (PCR) as a marker of thymic activity in post-HCT patients. Patients with active cGVHD had significantly lower expression of FOXP3 when compared with patients without cGVHD (P = .009) or healthy donors (P = .01). Moreover, patients with active cGVHD had significantly lower expression of FOXP3 when compared with patients without cGVHD (P = .009) or healthy donors (P = .01). Patients with or without cGVHD showed a significant decrease in TRECs compared with healthy donors (P < .001), supporting that thymic function is substantially impaired following allogeneic HSCT.

Therapeutic approaches: Based on studies showing the role of IL-2 as a growth factor for Tregs, clinical trials have been reported with use of low-dose IL-2 in cGVHD setting with promising response rates [24, 25]. Additional combinations of IL-2 with Tregs (NCT01937468) and photopheresis (NCT02340676) are ongoing to see if therapeutic response can be augmented above that seen with IL-2 monotherapy.

In addition to T cells, antigen-presenting cells (APCs), such as macrophages, dendritic cells, and B cells, play a critical role in initiating and propagating immune responses associated with cGVHD. Schlomchik et al. [11] demonstrated that host APCs are radio- and chemo-resistant post-conditioning regimens, and are critical for antigen presentation to incoming donor T cells, thereby proving the antigenic and co-stimulatory signals for T-cell activation and expansion leading to aGVHD. T-cell activation in the context of APC requires (1) antigen presentation with MHC class II molecules (MHC restriction) and (2) signal transduction via co-stimulatory molecules such as CD 80 and CD86. APCs are able to recognize the presence of microorganisms through the detection of conserved pathogen-associated molecular patterns (PAMP) and rapidly initiate tailored responses to these danger signals. Blockage of PAMP inducible co-stimulatory molecules such as CD40 or B7.1/B7.2 is effective in decreasing incidence of GVHD. She et al. [26] described the role of B cells primed for TLR9 (toll-like receptor 9), the response of which may play a role in pathophysiology of cGVHD. TLR9 expression has a significant correlation with expression of CD86 and CD80—furthermore, expression of these surface proteins was used as surrogate for TLR9 expression in this trial. A significantly greater percentage of B cells from and early cGVHD group (3–8 months post-HSCT; n = 19, 56.3 %) were capable of up-regulating CD86, compared to 6-month non-cGVHD controls (n = 9, 15.8 %; P = .0004) in response to PS-modified CpG. To confirm the B cell responses were mediated by TLR pathway, mRNA expression levels were checked in purified B cells by RT-PCR. There was a significant correlation (n = 12, r² = 0 .65, P = .002) between PS-modified CpG 2006 response and B cell TLR9 mRNA levels. Anderson et al. [27] further clarified the role of donor and host APCs in cGVHD. Both donor and host APCs can elicit cGVHD phenotype in setting of CD80/86 co-stimulation—in the absence of this co-stimulatory signal, no cGVHD developed. This process is CD4+ T cell mediated. Their findings show that donor APCs can cause late cGVHD in a CD4 cell dependent mechanism in setting of appropriate co-stimulatory signals. This finding provides additional therapeutic targets for prevention of cGVHD.

The role of stimulatory auto-antibodies in platelet-derived growth factor receptors (PDGFR) was studied by Svegliati et al. [31] in sclerotic cGVHD. Based on clinical evidence of agonistic antibodies toward PDGFR in patients with systemic sclerosis, they tested 39 post-HCT patients with cGVHD for presence of stimulatory auto-antibodies to the PDGFR. They detected the presence of stimulatory antibodies to the PDGFR in all patients with extensive cGVHD, but none in the patients without cGVHD. Their finding also supports the use of tyrosine kinase inhibitors (such as imatinib) as therapeutic agents in scleroderma by virtue of its anti-PDGFRA activity. Wechalekar et al. [32] studied, in 13 HCT recipients, the presence of auto-antibodies in rheumatoid factor (RF), antinuclear antibody (ANA), double-stranded DNA (dsDNA), antimitochondrial antibody, antismooth muscle antibody (AntiSm), antiendomysial, antireticulin antibodies, antithyroid peroxidase antibodies, and an extractable nuclear antigen screen. All the patients with antibodies had cGVHD, whereas none of the patients without cGVHD had any auto-antibodies (P = 0.025). Three of the patients (23 %) had only one autoantibody and three others (23 %) had more than one auto-antibody. ANA was positive in three patients (23.3 %), dsDNA in four patients (30.7 %), RF in one patient (7.6 %), and anti-smooth muscle in two patients (15.3 %). In the present study, auto-antibodies were detected predominantly in patients with the presence of cGVHD. They also appeared to be more frequent in an unmanipulated graft and less frequent in patients with a T-cell depleted allograft.