Therapeutic targeting of STAT3 pathways in pancreatic adenocarcinoma: A systematic review of clinical and preclinical literature

Background/Objectives Pancreatic ductal adenocarcinoma is a highly lethal disease with increasing incidence. Due to high resistance, chemo/radiotherapy has limited success in pancreatic cancer and only marginally prolongs patient survival. Therefore, novel biomarkers and therapeutic targets are needed. In the present review, we performed a comprehensive summary of therapeutic approaches targeting the GP130/JAK/STAT3 pathway. Methods We systematically reviewed the PubMed and Embase databases for preclinical and clinical studies, from inception to October 4, 2020, on drugs targeting the GP130/JAK/STAT3 pathway. Bias assessments and qualitative analyses were performed. Results Twenty-five preclinical and nine clinical trials were included in the review. All preclinical studies reported a favorable outcome in terms of pancreatic ductal adenocarcinoma progression. Futhermore, drugs targeting the GP130/JAK/STAT3 pathway were shown to be efficient chemosensitizers. However, high publication bias was assumed. In the clinical setting, bazedoxifene and itacitinib improved patient outcomes. Conclusion Preclinical studies strongly suggest significant efficacy of drugs targeting GP130/JAK/STAT3 in the treatment of pancreatic ductal adenocarcinoma and that these molecules are effective chemosensitizers. Though only a few trials have shown the efficacy in a clinical setting, the STAT3 pathway remains a promising drug target for future treatment of pancreatic ductal adenocarcinoma and may help overcome chemotherapy resistance.


Results
Twenty-five preclinical and nine clinical trials were included in the review. All preclinical studies reported a favorable outcome in terms of pancreatic ductal adenocarcinoma progression. Futhermore, drugs targeting the GP130/JAK/STAT3 pathway were shown to be efficient chemosensitizers. However, high publication bias was assumed. In the clinical setting, bazedoxifene and itacitinib improved patient outcomes.

Conclusion
Preclinical studies strongly suggest significant efficacy of drugs targeting GP130/JAK/ STAT3 in the treatment of pancreatic ductal adenocarcinoma and that these molecules are effective chemosensitizers. Though only a few trials have shown the efficacy in a clinical setting, the STAT3 pathway remains a promising drug target for future treatment of pancreatic ductal adenocarcinoma and may help overcome chemotherapy resistance.

Introduction
Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal disease with increasing incidence. In most cases, pancreatic cancer presents at an advanced stage, with only 20% of all cases undergoing surgical resection. In terms of prognostic outcomes for patients, pancreatic adenocarcinoma ranks last, with an overall 5-year survival rate of 2-9% [1,2]. Even though the management of pancreatic adenocarcinoma is evolving with the introduction of novel surgical techniques and medical therapies, only minor improvements in outcomes have been achieved. Due to high resistance, chemotherapy and radiotherapy have limited success in metastatic PDAC and only marginally prolong patient survival [3]. Current treatment options for metastatic PDAC are modified FOLFIRINOX/FOLFIRINOX or nab-paclitaxel and gemcitabine in patients with good performance status, and gemcitabine with or without a second agent for those with poor performance status [4]. Most recently, trials studying the update of immunotherapy in PDAC were negative except in a subgroup of adenocarcinoma with microsatellite instability [5].
Considering the lack of effective treatment, the identification of novel biomarkers and therapeutic targets is fundamental to developing new treatment strategies and improving clinical outcomes. Recent studies suggest that signaling pathways involving STAT3 play a key role in tumorigenesis, progression and drug resistance in several human malignancies such as leukemia, lymphomas as well as solid tumors such as hepatocelullar carcinoma, esophageal, lung, prostate, bladder and breast cancer [6,7]. Animal models of PDAC have shown that STAT3 is an important regulator of stem cell self-renewal and cancer cell survival [8,9]. Upregulation of STAT3 has been shown to promote the development of PDAC from pancreatic intraepithelial neoplasia [10,11], as well as pro-metastatic niche formation in the liver [12]. Furthermore, STAT3 has been shown to mediate resistance to chemotherapy and to be associated with adverse outcomes following resection of PDAC with curative intent [13][14][15].
Accordingly, pathways involving STAT3 appear to be promising drug targets for the treatment of PDAC. In particular, IL-6 has been shown to be a potentially efficient therapeutic approach for overcoming chemotherapy resistance. The purpose of this study was to provide a comprehensive summary of therapeutic approaches targeting the GP130/JAK/STAT3 pathway in pancreatic adenocarcinoma through a systematic qualitative review of the literature.

Methods
This systematic review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [24]. Studies were identified by searching PubMed using the following search terms: Embase was searched using the following search query:  The electronic search was supplemented by a manual search of the reference lists of relevant articles to identify any studies that may have been missed in the database searches. The original database search was performed on May 5, 2020. The electronic search was updated on October 4, 2020.
Inclusion criteria were defined as all trials studying the pharmacological targeting of the GP130-related cytokine/JAK/STAT3 pathway in pancreatic cancer, including studies on animal models or cell cultures. Only studies with an English abstract were included. Reviews, comments, and conference or meeting abstracts were excluded from the analysis. No restrictions on publication date or publication status were imposed.
After exclusion of duplicates, records identified from the literature search were screened for eligibility independently by the two main authors using the title and abstract in an unblinded manner. Disagreements between the reviewers were resolved by consensus. The full-text of articles meeting the inclusion criteria was assessed by the two main authors and reevaluated for the inclusion criteria. Disagreements were, again, resolved by consensus.
We extracted data using a previously prepared extraction form. The information from each included study on the study design, characteristics of analyzed subjects or trial participants, characteristics of the pharmacological agent studied, type of outcome measures, and outcomes was tabulated. On this basis, we performed a qualitative data synthesis.
We performed a quality assessment of clinical trials according to the ROB tool, which was adapted to match non-randomized clinical trials [25]. For preclinical studies, we used the SYR-CLE's risk of bias tool, which was adapted to match in vivo and in vitro studies [26]. Results were displayed in an analogous fashion as suggested by Higgins et al for systematic reviews of interventions [25]. Bias assessment was conducted for every study by two independent assessors and disagreements resolved by consensus.
Due to the nature of this study, approval from the local Ethics Committee was not required.

Study selection
Our search identified 756 records through the database searches (Embase, Pubmed) and the manual search of the reference lists of relevant articles. Initial screening excluded 689 records, including 145 duplicates. The remaining 67 articles were assessed based on the full text, 29 of which were found to be ineligible due to absence of a tested pharmacological substance or the absence of GP130-related cytokine/JAK/STAT3 pathway targeting. A summary of the study selection process is provided in Fig 2. Ultimately, 38 studies were included in the review, including 4 ongoing trials. All included studies were published in English and no unpublished data were included. No other studies were identified through the electronic search update on October 4, 2020. Table 1 shows the risk of bias assessment for the preclinical studies. Preclinical studies had strong limitations to rigorous bias assessment because few provided sufficient details regarding selection and performance bias. Study protocols were not published beforehand, so a comparison between intended interventions and published interventions was not possible. In animal trials, few studies explicitly stated a randomization process for treatment groups, and treatment results were often assessed manually with semi-quantitative methods. This lack of reporting makes it difficult to accurately determine the risk of bias of the preclinical studies. However, more details were available on the risk of attrition bias, reporting bias, and other bias.

Bias assessment
The quality assessment of the included clinical trials is provided in Table 2. The overall quality of the studies was good, with only one study presenting high risk of selection bias.

Clinical studies
As summarized in Table 4, nine of the studies were clinical trials including 880 individuals and assessing 5 drugs. One study performed a retrospective analysis of bazedoxifene, an inhibitor of the IL-6/IL-6R/GP130 complex, in patients with pancreatic (n = 5) or gastric adenocarcinoma (n = 2), showing biological tumor marker reduction in 80% and disease regression on PET-CT in 60% of cases [51]. Icatinib, a selective JAK1 inhibitor, was tested in combination with nab-paclitaxel and gemcitabine, showing a synergistic effect with an overall response rate of 24% with an acceptable safety profile in a phase 1b/2 study [52]. However, this study was terminated early due to negative phase 3 results for JAK1/2 inhibitor ruxolitinib [55]. Momelotinib, a JAK1/2 inhibitor, resulted in a partial response in 28% of patients with previously untreated metastatic PDAC (n = 25) in a phase 1 study. However, no significant difference was reported from treatment with paclitaxel and gemcitabine [53]. Ruxolitinib, a JAK1/2 inhibitor, has been investigated in phase 1b, 2, and 3 clinical trials in combination with capecitabine, gemcitabine, and paclitaxel, revealing no significant difference in overall survival or progression-free survival in patients with PDAC [54][55][56]. Finally, phase 2 and 3 studies have been performed assessing tipifarnib, an inhibitor of STAT3 phosphorylation that showed no singleagent antitumor activity and no difference in overall survival in combination with gemcitabine  [57][58][59]. Four ongoing clinical trials were found, involving tocilizumab, an anti-IL6Rα antibody with favorable results in preclinical studies [31,33], and napabucasin, a STAT3 inhibitor that is also under investigation in colorectal cancer [63].

Discussion
The present systematic review of 25 preclinical studies and 9 clinical trials revealed a good overall effect of the investigated drugs targeting the GP130/JAK/STAT3 pathway in the treatment of PDAC. Table 5 summarizes the outcome and the state of research for each assessed drug. Favorable outcomes have been reported for all 20 drugs investigated in a preclinical setting. Even though these substances appear promising in the treatment of PDAC, only five of these drugs have been investigated in clinical trials. Favorable outcomes and acceptable toxicity profiles have been found in studies investigating bazedoxifene and itacitinib [51,52]. Notably,

PLOS ONE
Targeting STAT3 in pancreatic cancer bazedoxifene is already approved for the treatment of osteoporosis [64], and itacitinib has been shown to have great potential in recent clinical trials studying the treatment of connective tissue diseases and graft-versus-host disease, among others [65][66][67].
Even though the PDAC tumor micro-environment (TME) has been shown to be a promising target for improving PDAC treatment, none of the included studies in this systematic review examined the influence of the analyzed substances on stromal or immune cells.

PLOS ONE
Targeting STAT3 in pancreatic cancer The TME plays an important role in tumorigenesis and chemoresistance by close interaction with tumor cells. Furthermore, TME has been shown to be highly immunosuppressive, promoting immune evasion, hence sustaining tumor progression [18][19][20]. Immunotherapy, so far, has not demonstrated substantial clinical improvement as single agent in the treatment of PDAC [5]. Therefore, strategies simultaneously targeting PDAC tumor cells as well as different immune checkpoints might be needed. The interactions of PDAC tumor cells and different cells within the TME such as CAFs, MDSCs, TAMs, are mediated through GP130/JAK/STAT3 pathway [11,[17][18][19][20]68]. STAT3 inhibition might thus have consequences in shaping TME towards anti-tumor phenotype by acting on both immune and tumor cells [18][19][20]. In combination with chemotherapeutic agents and immunotherapy, it might significantly increase therapeutic efficacy in the treatment of PDAC.
Recent studies have shown the important role of the STAT3 pathway in tumorigensis, as well as the STAT3-mediated resistance to chemotherapy in in vivo models of PDAC [8][9][10][11][12][13][14][15]. The results from the preclinical trials presented in this review confirmed the importance of the GP130/JAK/STAT3 pathway in PDAC and its role as a possible drug target. Furthermore, several of the studies showed a synergy between the investigational drug and chemotherapy, such as gemcitabine, paclitaxel, 5-fluorouracil, and oxaliplatin [31, 33, 38, 40, 41, 45, 49, 53-55, 57, 60]. However, to the best of our knowledge, drugs targeting GP130/JAK/STAT3 have never been studied as chemosensitizers in addition to the currently emerging FOLFIRINOX regimen [4]. Even though some promising outcomes have been shown in clinical trials [51,52], several studies were terminated prematurely due to high progression rates and futility. This may reflect the difficulty showing a significant benefit in patients presenting with PDAC, as it is known to be a highly lethal disease that is often diagnosed at an advanced stage and has a poor prognosis with an overall 5-year survival rate of 2-9% [1,2].
The discrepancy between preclinical and clinical data may also result from the fact that, in contrast to the preclinical studies, the clinical trials did not verify the activation of the STAT3 pathway in PDAC. The benefit of targeted GP130/JAK/STAT3 therapy may be increased by selecting patients with previously known STAT3 pathway activation in PDAC cells.
The present systematic review included all preclinical and clinical trials of drugs targeting the GP130/JAK/STAT3 pathway. Furthermore, we searched for ongoing, unpublished trials, leading to a thorough analysis of the current state of research. However, because all published preclinical studies reported a positive outcome, we suspect that several negative studies may not have been published and concluded relevant publication bias, leading to an overestimation of the effect of GP130/JAK/STAT3-targeting drugs in the treatment of PDAC in the preclinical setting. Furthermore, the substantial heterogeneity among the preclinical and clinical studies did not allow a quantitative analysis or measurement of the effect size.

Conclusion
Preclinical studies strongly suggest significant efficacy of drugs targeting GP130/JAK/STAT3 in the treatment of PDAC and that these molecules are effective chemosensitizers, possibly through simultaneous effect on tumor cells and TME. Though only a few trials have shown the efficacy in a clinical setting, the GP130/JAK/STAT3 pathway remains a promising drug target for the development of future treatments for PDAC and may help overcome chemotherapy resistance.