Pancreatic ductal adenocarcinoma (PDAC) is a crazy "killer", with a 10-year survival rate of about 1% in patients with PDAC [1].
Over the past 20 years, targeted therapy and immunotherapy have been progressive, rewriting the treatment paradigm for many cancers; unfortunately, the classical glandular subtype has some response to current clinical treatment options [2,3], while the basal mesenchymal subtype of PDAC has little response to standard chemotherapy and immunotherapy.
There is a reason why PDAC is so difficult to treat. PDAC has been found to have a lower tumor mutation burden and produce fewer immunogenic neoantigens, which makes it difficult for T cells to recognize cancer cells; in addition, the PDAC tumor microenvironment (TME) is immunosuppressive, making it difficult for tumor-infiltrating lymphocytes (TILs) to be recruited to the tumor site [4]. The good news is that a small number of cases of PDAC characterized by high levels of T cell infiltration have recently been reported, and this feature has been associated with prolonged overall survival in patients [5-7].
Based on the above research results, it is not difficult to find that the combination of therapy that can improve the tumor microenvironment and immunotherapy may be an effective way to treat PDAC.
Recently, a team of Dieter Saur from the German Cancer Research Center in Heidelberg, Germany, and the Translational Cancer Research Division of the German Cancer Society published important research results in the journal Nature Cancer [8].
Through systematic high-throughput screening, they found that the MEK inhibitor trametinib combined with the polykinase inhibitor nintedanib can lead to cancer cell cycle arrest and cancer cell death, and also promote cytotoxicity and intratumoral infiltration of effector T cells. That is, the combination of the two can not only kill cancer cells, but also reprogram the tumor microenvironment.
More importantly, trimetinib plus nidanib also sensitizes refractory interstitial PDAC to PD-L1 inhibitors. Overall, this study opens up new therapeutic pathways for refractory interstitial PDAC.
It is well known that more than 90% of patients with PDAC develop mutations in KRAS. Although drugs that target KRAS mutations have been used to treat other cancers, so far, these drugs have not been effective in treating KRAS-mutated PDAC. It seems that the breakthrough may have to be found downstream of the KRAS pro-cancer pathway.
Downstream of carcinogenic KRAS, the RAF-MEK-ERK pathway plays a central role in tumorigenesis. MEK inhibitors (MEKi) have shown good anti-cancer effects in RAS-mutated melanoma and lung cancer, but unfortunately, in PDAC patients, MEKi has failed.
Saur's team's previous research found that mutations and constant amplification of the carcinogenic KRAS gene drive the early occurrence and metastasis of PDAC. At the same time, they also found that KRAS mutants with highly aggressive mesenchymal PDAC had the highest levels of expression [9].
After learning that KRAS mutant levels have a strong influence on the PDAC phenotype, Saur's team had new inspiration. They want to develop a combination therapy that can both block KRAS-driven intrinsic signal transduction in tumor cells and reprogram TME to improve the effectiveness of mesenchymal PDAC treatment.
To achieve the above goals, Saur's team first systematically explored the relationship between inhibition of the typical KRAS–RAF–MEK–ERK pathway and the benefits of PDAC treatment.
They used MEKi (trimetinib) to treat a group of primary patient-derived PDAC cells and conventional human PDAC (hPDAC) cell lines. Unexpectedly, the classical glandular subtype hPDAC is highly sensitive to trametinib (Figure 1b).
Due to the lack of hPDAC cells representing the most undifferentiated, most aggressive, and intact mesenchymal morphology in clinical specimens, the researchers expanded the screening by isolating primary PDAC cells (mPDAC) from mouse pancreatic cancers expressing KrasG12D and repeating the experiment. KrasG12D expression was highest in mPDAC cells of the mesenchymal subtype compared to mPDAC of the classical glandular subtype (Figure 1c). As with hPDAC, it is predominantly classical glandular subtype mPDAC cells that are sensitive to trametinib, while almost all mesenchymal subtypeS PDAC are resistant to trammetinib (Figure 1d).
Subsequently, Saur's team demonstrated through in vitro and in vitro experiments that the complete and sustained disruption of classical KRAS downstream signals using MEKi or gene knockout was not enough to inhibit the growth of mesenchymal subtype PDAC tumors.
Figure 1: Survival curves of patients with surgical resection of a, G1–G2, or G3–G4 tumor grades. b. Percentage of cell viability of 10 nM trammetinib in hPDAC cell lines. c. Expression of classical and mesenchymal PDAC intermediate gene-specific KrasG12DmRNA. d. Cell viability of mPDAC cells at 10 nM trammetinib.
Next, Saur's team conducted a systematic, high-throughput screening of the compounds in combination to identify drugs that had synergistic effects with trametinib.
They were screened with trimetinib, each in combination with 418 drugs, in human and mouse PDAC representing the classical glandular subtype and the mesenchymal subtype of KRAS mutations. After a large number of cell experiments, it was found that trametinib and nidanib (T/N) have a significant synergistic effect in the treatment of PDAC of the intercosemic subtype in humans and mice. Nidanibu is a clinically approved RTK inhibitor and one of the most popular drugs for the treatment of mesenchymal PDAC.
In order to find the direct targets of trimetinib and nidanib, the researchers further studied 6 mPDAC of classical glandular subtypes and mesenchymal subtypes, and found that trimetinib can selectively inhibit MEK1/2, while nidanib has a wide range of targets, mainly concentrated in RTK and cell surface receptors.
Subsequently, to identify pathways that mediate T/N response correlation, Saur's team analyzed changes in the phosphoproteome. In the mesenchymal subtype of PDAC, the activity of a number of important cancer-related pathways is reduced, such as the cell cycle regulator cyclin-dependent kinase 2 (CDK2), cyclin D and cyclin E, PP2A and IER3, ERBB2, mTOR, and KIT downstream signaling that regulate PI3K/AKT signaling, as well as RAF-dependent and non-dependent ERK1/2 activation. These results suggest that the PDAC of the mesenchymal subtype relies on a wide range of RTK-driven signal inputs.
These findings suggest that multiple kinase targets are needed to achieve effective treatment in kras-mutant mesenchymal subtype PDAC!
To further decipher key genes for T/N synergy, Saur's team used whole-genome hybrid screening in mouse PDAC cells of three mesenchymal subtypes, as well as CRISPR-based gene editing technology screening. Of the 53 nidanib targets identified in previous experiments, 15 were found to be functionally related to trimetinib.
In addition, through CRISPR gene editing technology, they also found that the combined deletion of multiple targets led to the sensitivity of PDAC of the mesenchymal subtype to trametinib, of which the combined deletion of Prkaa1, FGFR1 and Map2k5 was the most significant. This once again confirms the need for a wide range of targets to effectively and comprehensively treat krass-mutant mesenchymal PDAC.
Exciting in vitro experimental results have driven researchers to construct mouse PDAC in situ transplant models of classical glandular subtypes and mesenchymal subtypes to explore the efficacy of in vivo combination therapy. Experimental results showed that the combination T/N therapy effectively inhibited the growth of PDAC tumors of the mesenchymal subtype, significantly reduced the tumor volume by about 40%, and doubled the survival rate of mice (Figures 2b-d).
Figure 2 a Kaplan–Meier curve compares the survival of classical and mesenchymal in situ PDAC models. b MRI evaluates changes in tumor volume in classic glandular subtypes and mesenchymal PDAC after 1 week of treatment. c Representative MRI of mice in the control and T/N treatment groups. d Kaplan–Meier survival curves for classical and mesenchymal in situ models.
Happily, this is the first combination therapy to be effective against kras mutant gene amplification-driven mesenchymal PDAC! As for the mechanism behind it, Saur's team found that T/N treatment significantly increased the infiltration of T cells into mesenchymal PDAC, transforming "cold tumor" PDAC into an immune-responsive "hot tumor."
At the same time, they also found an interesting phenomenon. Tumor sections of mesenchymal PDAC showed that increased CD8+ T cell infiltration was most pronounced around the blood vessels, which they thought might be due to the remodeling of the blood vessels with T/N combination therapy. In contrast, PDAC tumors of the classical glandular subtype exhibit features of immune rejection, with only moderately enriched T cells at the tumor margins (Figures 3a-d). It is shown that the combination of T/N can only reprogram the PDAC of the mesenchymal subtype, and the effect on the classical subtype is not obvious.
Figure 3 a proportion of adaptive immune cell populations in T/N-treated mouse tumors. b Flow cytometric analysis of tumor CD4+ and CD8+ T cells treated with T/N for 1 week. c Representative images of IHC staining of tumor sections in an in situ transplanted mesenchymal model for 1 week of T/N treatment with CD3+ and CD8+ T cells. Representative image (green) of d CD3+ cell stained tissue sections.
Subsequently, in order to study the role of T cells in the treatment response, Saur's team constructed a mouse model of T cell deletion, and found that the loss of T cells weakened the treatment effect of T/N, and the survival time of mesenchymal PDAC mice was also reduced, but compared with the control group, the survival time of tumor-bearing mice was extended to a certain extent (Figure 4). It is suggested that T/N onset is not mediated by T cells alone, but may also be related to TME reprogramming and direct drug effects on tumor cells.
Figure 4 Kaplan-Meier survival curves of CD3 knockout and in situ transplantation of C57BL/6WT mice for classic (top) and mesenchymal (bottom) PDAC
Based on the above major findings, since T/N can promote the death of mesenchymal PDAC cancer cells and reshape their microenvironment, and recruit cytotoxic T cells to infiltrate the tumor site, does this mean that mesenchymal PDAC can also be sensitive to IDB treatment?
With the above question in mind, Saur's team validated the T/N combined PD-L1 inhibitor on a mouse mesenchymal PDAC model. The experimental results showed that T/N combined with PD-L1 inhibitor therapy could make the tumor suppression rate as high as about 80% and improve the survival rate of mesenchymal PDAC mice, of which the mid-term survival was 10.5 days longer than that of the T/N treatment group and 30.5 days longer than that of the control group. In contrast, combining PD-L1 inhibitors on a T/N basis had little effect on the classical glandular subtype PDAC. In addition, neither subtype responded to PD-L1 inhibitors alone (Figures 5a, c).
Figure 5 a Waterfall plot shows the change in tumor size of classical glandular subtype and mesenchymal PDAC versus T/N combined with PD-L1 inhibitor therapy after 1 week of treatment. c Kaplan–Meier survival curve for T/N plus PD-L1 inhibitor therapy with classical glandular subtype and mesenchymal PDAC.
Finally, in order to comprehensively and objectively study TME changes induced by T/N treatment, and to mechanically decipher the drug's effect on the PDAC tumor microenvironment of classical glandular subtypes and interstitial subtypes, Saur's team also performed single-cell RNA sequencing (scRNA-seq) for tumor tissue.
They found that CD4+ and CD8+ T cells with immature T cell gene expression characteristics in T/N-treated mesenchymal PDAC were significantly reduced, while mature T cells with functional cytotoxicity, effectivity, and memory gene expression characteristics were significantly increased (Figure 6b-c). After adding PD-L1 inhibitors to T/N, cytotoxic T cells and effector T cells increase further, accounting for almost 75% of all T cells (Figure 6b). In addition, in mesenchymal PDAC, the T/N combination specifically induces secretion of CXCL12, CXCL16 and TNFSF12, while the expression of CCL2, CSF1 and LGALS9 is down-regulated.
Figure 6 a Left: UMAP plot showing the expression of CD3g, CD4, and CD8a marker genes in the entire population of T cells identified by scRNA-seq in classical and mesenchymal PDAC. Medium figure: UMAP profile of T cells (yellow) and T cells (blue) of interstitial PDAC for all treatment and control groups. Right panel: The UMAP plot shows 6 T cell subsets identified by scRNA-seq. b Cell proportions by treatment conditions and PDAC subtypes, by scRNA-seq analysis of T cell clusters annotated in a. c Expression heat map of selected genes in selected T cell clusters of classical and mesenchymal PDAC.
In summary, T/N treatment can activate the TME of mesenchymal PDAC, which is beneficial to ICBC therapy.
Overall, Dieter Saur's team's research, through single-cell RNA sequencing, CRISPR screening, and immunophenotyping, found that the combination of trimetinib and nidanib can reshape the tumor microenvironment of mesenchymal PDAC and promote the infiltration of cytotoxic T cells and effector T cells within the tumor, thereby sensitizing mesenchymal PDAC to PD-L1 inhibitors.
The results of this study open up new ideas and avenues for the treatment of mesenchymal PDAC with highly invasive and refractory KRAS mutations, and this combination therapy may be applied to other refractory tumors with poor response to immunotherapy. In addition, it also suggests that in the development of drugs that are screened in vitro for clinical practice, TME should be considered to remodel to enhance the anti-tumor effect.
bibliography:
[1] Quaresma M, Coleman M P, Rachet B. 40-year trends in an index of survival for all cancers combined and survival adjusted for age and sex for each cancer in England and Wales, 1971-2011: a population-based study[J]. Lancet, 2015, 385(9974): 1206-18.
[2] Aung K L, Fischer S E, Denroche R E, et al. Genomics-Driven Precision Medicine for Advanced Pancreatic Cancer: Early Results from the COMPASS Trial[J]. Clin Cancer Res, 2018, 24(6): 1344-1354.
[3] S N K, Wilson G W, Grant R C, et al. Morphological classification of pancreatic ductal adenocarcinoma that predicts molecular subtypes and correlates with clinical outcome[J]. Gut, 2020, 69(2): 317-328.
[4] Morrison A H, Byrne K T, Vonderheide R H. Immunotherapy and Prevention of Pancreatic Cancer[J]. Trends Cancer, 2018, 4(6): 418-428.
[5] Balachandran V P, uksza M, Zhao J N, et al. Identification of unique neoantigen qualities in long-term survivors of pancreatic cancer[J]. Nature, 2017, 551(7681): 512-516.
[6] Chen D S, Mellman I. Elements of cancer immunity and the cancer–immune set point[J]. Nature, 2017, 541(7637): 321-330.
[7] Ino Y, Yamazaki-Itoh R, Shimada K, et al. Immune cell infiltration as an indicator of the immune microenvironment of pancreatic cancer[J]. Br J Cancer, 2013, 108(4): 914-23.
[8] Falcomatà C, B rthel S, Widholz S A, et al. Selective multi-kinase inhibition sensitizes mesenchymal pancreatic cancer to immune checkpoint blockade by remodeling the tumor microenvironment[J]. Nature Cancer, 2022.
[9] Mueller S, Engleitner T, Maresch R, et al. Evolutionary routes and KRAS dosage define pancreatic cancer phenotypes[J]. Nature, 2018, 554(7690): 62-68.
Responsible editor 丨IoTalker