Pancreatic cancer remains a poor-outcome disease with mortality rates nearly identical to incidence rates. Autophagy is a key metabolic process involved in stress resistance both intrinsically in tumor cells, and extrinsically in the tumor microenvironment (TME).
1. Key Features of Pancreatic Ductal Adenocarcinoma (PDAC)
PDAC is the most frequent type of pancreatic cancer (90%). It remains a devastating disease with poor prognosis and limited efficiency of commonly available therapies
[1][2]. PDAC is the fourth leading cause of cancer-related deaths in Western societies with a 5-year overall survival rate of only 10%
[3][4]. Due to its late diagnosis, high metastatic capacity, aggressive local progression, and increase in incidence, PDAC is predicted to become the second cause of cancer deaths by 2030 in the US
[4][5][6].
PDAC is a malignant epithelial tumor that arises from the exocrine portion of the pancreas and is mainly found in the head of the organ. Microscopically, PDAC consists of atypical tubular glands, partially formed or poorly formed glands with solid areas, depending on whether it is a well, moderate, or poorly differentiated neoplasm, respectively
[7]. Importantly, PDAC develops from macroscopic and microscopic precursor lesions designated intraductal papillary mucinous neoplasms (IPMNs) and pancreatic intraepithelial neoplasias (PanINs) (
Figure 1), respectively, the latter being believed to develop and progress asymptomatically over several decades
[7][8]. Despite the ductal appearance of PanINs, it has been shown that acinar cells are at the origin of these lesions. Under specific conditions such as pancreatitis or genetic alterations, acinar cells can transform into duct-like structures, termed acinar–ductal metaplasia (ADM)
[9]. ADM frequently progress to the well-characterized PanINs, that evolve to pancreatic tumors over time (
Figure 1). Thus, ADM lesions can be considered as the main origin of PDAC
[10].
Figure 1. Schematic representation of Pancreatic Ductal Adenocarcinoma (PDAC) carcinogenesis, the main genetic events involved, and key metabolic features. PDAC is a malignant epithelial neoplasm that arises from the exocrine portion of the pancreas, mainly from the acinar cells. Under severe stress conditions such as pancreatitis, acini can transform into duct-like structures, termed acinar–ductal metaplasia (ADM). ADM frequently progress to the well-characterized pancreatic intraepithelial neoplasias (PanINs), defined as mucinous-papillary proliferations with a ductal appearance. These lesions can be classified into low or high grade PanINs based on the degree of architectural and nuclear atypia. Genomic mutations of pancreatic cancer predominate in four genes, the KRAS oncogene and the tumor suppressor genes CDKN2A, TP53, and SMAD4. KRAS oncogenic mutations are nearly ubiquitous in PDAC and occur early in carcinogenesis, and more importantly, these point mutations drive constitutive KRAS activation thus maintaining cell proliferation and survival. Subsequent mutations in the tumor suppressor genes are present later further contributing to disease progression. Besides genetic alterations, PDAC is characterized by a prominent fibrotic stroma (desmoplasia), consisting of abundant extracellular matrix (ECM) and stromal cells in the tumor microenvironment (TME). This TME is very heterogeneous in terms of the variety of stromal cell subtypes, which includes cancer-associated fibroblasts (CAFs, in yellow), immune cells such as lymphocytes (blue) and macrophages (green), stroma-associated pancreatic stellate cells (PSCs, in gray), among others. This prominent TME exerts high levels of solid and fluid pressure, and compression of vasculature, limiting oxygen and nutrient availability during PDAC progression. Notwithstanding, pancreatic cancer cells are well adapted to these adverse conditions by several mechanisms, including autophagy, which shows both pro- and anti-tumorigenic effects depending on the context. In PDAC, autophagy can prevent cancer initiation at early steps of the disease, and in established tumors, autophagy supports PDAC growing and maintenance by different mechanisms.
2. Therapeutic Approaches Using Autophagy Inhibitors in PDAC
Autophagy has been shown to be elevated in PDAC and it is implicated in resistance to both cytotoxic chemotherapy and targeted therapy
[11]. Within this framework, the essential role of autophagy in promoting pancreatic cancer has made it a promising therapeutic target in PDAC
[12][13]. Clinical interventions to manipulate autophagy in cancer therapy are already under investigation, with the vast majority focused on autophagy inhibition
[14]. The most widely employed compounds that inhibit the last stage of autophagy are Chloroquine (CQ) and its derivative Hydroxychloroquine (HCQ), Bafilomycin A1 (BafA1), and lysosomal protease inhibitor cocktails. CQ and HCQ are synthetic 4-aminoquinolines initially developed for malaria disease and thus are the only autophagy inhibitors approved by the Food and Drug Administration (FDA)
[15][16]. Therefore, their effects on several cancers, including pancreatic, have been deeply studied as explained below.
2.1. Preclinical Trials
Since autophagy is a complex and context-dependent process in PDAC as in other cancers, its targeting has to be well supported by preclinical data regarding the role and status of autophagy in a particular cancer
[17]. In this regard, several preclinical studies support the idea that autophagy inhibition may improve PDAC treatment in a clinical setting. A large number of in vitro assays and in vivo studies using genetically-engineered mouse models (GEMMs) and patient-derived xenograft (PDX) have been conducted
[14]. Albeit genetic interventions in autophagy are widely used in cancer research, pharmacological inhibition is more kinetically controllable, and is the most frequently employed approach in preclinical studies
[15].
Due to their elevated autophagy, PDAC cells are highly sensitive to autophagy inhibition. Accordingly, CQ markedly decreased the proliferation and anchorage-independent growth in a panel of PDAC cell lines but had minimal impact on other cancer cells with low basal autophagy flux
[18]. Moreover, autophagy inhibition with CQ has a potent impact in vivo since it can both eradicate PDAC growth and prolong survival in subcutaneous and orthotopic xenografts and in a KRAS-driven GEMM
[18]. Importantly, in the subcutaneous model, tumors seemed to have a sustained complete regression in half of the CQ-treated mice
[18]. Interestingly, in a large panel of PDX with
TP53 mutations, HCQ treatment attenuated the growth of most xenografts, showing the clinical relevance of autophagy inhibition independent of p53 status
[19].
The impact of autophagy inhibition with CQ has also been tested in pancreatic CSCs, resulting in decreased CSCs populations, sphere formation, and resistance to gemcitabine in vitro and in vivo. Consequently, autophagy blockade sensitized pancreatic CSCs to gemcitabine and enhanced its efficacy against PDAC. Clearly, in this work a high coexpression of LC3 and ALDH1 correlated with poor prognosis, and this was mediated by osteopontin (OPN)/NF-κB signaling
[20]. Although HCQ has shown improvement in PDAC studies, there has been controversy about its off-target effect, particularly the inhibition of other aspects of lysosomal scavenging. Therefore, Yan and colleagues developed an inducible mouse model (Atg4B mutant) that allowed the acute and reversible inhibition of autophagy. Excitingly, this inhibition caused significant tumor regression, being a result of both tumor cell-intrinsic and host effects
[11].
Given that KRAS mutations are almost ubiquitous to PDAC, the National Cancer Institute has identified the development of anti-KRAS therapies as one priority for pancreatic cancer research. One approach is to target KRAS-dependent metabolic functions, including autophagy. Paradoxically, ERK inhibition in PDAC impaired glycolysis and mitochondrial metabolism, leading to increased autophagy. However, ERK1/ERK2 inhibition in association with CQ enhanced the autophagy-inhibitory activity of the latter in vitro, blocking tumor progression and extending survival using HCQ in KRAS-mutant pancreatic subject-PDX
[21]. In a similar way, inhibition of MEK1/2 with Trametinib in combination with CQ resulted in a synergistic anti-proliferative effect against PDAC cell lines and induced regression of xenografted pancreatic tumors
[22].
Immunotherapy has not been successful in pancreatic cancer, being largely refractory to immune checkpoint inhibitors (ICI)
[23]. However, systemic autophagy inhibition with CQ, as well as tumor-specific autophagy inhibition, sensitizes PDAC to immune blockade. This occurs via restoring MHC-I surface levels and enhancing anti-tumor CD8
+ T cell responses in immunocompetent host mice
[24]. Therefore, targeting autophagy may promote the efficacy of immunotherapy and overcome immunotherapy resistance
[25].
It is important to mention that CQ has been reported to present anti-tumor effects independent of autophagy, as shown by Eng et al., where Atg7-deficient and -proficient cells were equally sensitive to the antiproliferative effect of CQ
[26]. This demands appropriate genetic stratification parameters to predict efficacy towards this compound. Finally, besides autophagy inhibitors, it will be important to determine the efficacy of autophagy inducers in combination with chemotherapeutics, immunotherapy, or MAPK inhibitors
[27].
2.2. Clinical Trials
The development of clinical trials is a result of a vast number of preclinical studies, some of them with promising results, such as those works showing complete tumor regression of PDAC tumors in different mouse models. Due to its clinical availability and well-recognized activity as an autophagy inhibitor, HCQ has been tested in several clinical trials in association with chemotherapy or radiotherapy
[28].
In pancreatic cancer, most clinical trials targeting autophagy used HCQ in combination with standard chemotherapies or MAPK inhibitors. In these trials, besides the objective of improving patient survival, other clinical endpoints include the serum carbohydrate antigen (CA) 19-9 biomarker
[16][29], histopathologic response
[16], and immune response
[16][30], important for assessing high risk tumors, margins after surgery, and immunomodulatory effects, respectively. Targeting autophagy could potentially enhance immunotherapy, and clinical trials using HCQ in combination with immunotherapy are ongoing to treat patients with different types of cancer
[25]. In the case of pancreatic cancer, a sole clinical trial is published using a combinatorial strategy of HCQ/gemcitabine/nab-paclitaxel/avelumab (NCT03344172), however, it was terminated due to suspected serious adverse effects related to the treatment.
Importantly, it is recognized that CQ and HCQ can have several off-target effects, thus Zeh and colleagues analyzed autophagic markers in resected PDAC tumors, finding the sequestosome protein SQSTM1/p62 accumulation associated with HCQ treatment
[16]. Interestingly, in this study, no difference was found for the common recycled autophagy marker LC3B II/Atg8 between controls and treated tumors. Still, in different types of solid cancers, the ability of HCQ to inhibit autophagy, together with its safety, has been demonstrated supporting its use for treating cancer
[31].
Table 1 recapitulates the completed or ongoing clinical trials using CQ or HCQ as autophagy inhibitors in the treatment of pancreatic cancer.
Table 1. Clinical trials targeting autophagy in pancreatic cancer.