Skip to main content
NCBI home page
ESMO Gastrointestinal Oncology logoLink to ESMO Gastrointestinal Oncology
. 2024 Mar 15;4:100042. doi: 10.1016/j.esmogo.2024.100042

Exploring the differences in the tumor microenvironment and immuno-oncologic targets in pancreatic ductal adenocarcinomas (PDAC) according to KRAS mutational status

EB Faber 1,2, Y Baca 3, J Xiu 3, P Walker 3, G Manji 4, S Gholami 5, A Saeed 6, A Prakash 1, GP Botta 7, D Sohal 8, HJ Lenz 9, AF Shields 10, C Nabhan 3, W El-Deiry 11, A Seeber 12, V Chiu 13, J Hwang 1, E Lou 1,∗
PMCID: PMC12836609  PMID: 41648036

Abstract

Background

The majority of pancreatic ductal adenocarcinomas (PDACs) are driven by mutant (mt) KRAS. How mt KRAS and co-driver mutations affect the immune cell (IC) landscape of PDAC remains uncertain. Herein, we characterize the types of IC in the PDAC tumor microenvironment (TME) and the prevalence of immuno-oncologic (IO) biomarkers by genomic and transcriptomic analysis in the context of KRAS status.

Materials and methods

4142 PDAC and 3727 colorectal cancer (CRC) cases with KRAS mt were analyzed using next-generation DNA sequencing, immunohistochemistry, and whole-transcriptome RNA sequencing. Microsatellite instability and deficiency in mismatch repair (MSI-H/dMMR) and tumor mutational burden (TMB) were also assessed.

Results

We found KRAS mt in 81% of PDAC, with the most common variant being G12D in PDAC, and fewer cases of KRAS mt were co-expressed with the predictive IO marker MSI-H/dMMR than KRAS-wild-type (wt). However, KRASG12D, KRASG12V, and KRASQ61 mutations had significantly lower TMB than KRAS wt tumors in PDAC. The IC environment of KRAS mt PDAC showed significant differences in nearly all IC types; a similar pattern was observed in CRC but was less pronounced.

Conclusions

Therapeutic IO targets like programmed death-ligand 1 are enriched in pancreatic adenocarcinoma cases harboring specific targetable variants of KRAS mt PDAC. Better understanding of the TME could lead to tailored immunotherapeutic strategies to overcome these barriers in KRAS mt PDAC, possibly in combination with molecularly targeted treatment strategies.

Key words: pancreatic cancer, pancreatic adenocarcinoma, KRAS, genomics, tumor microenvironment

Highlights

  • •

    We analyzed 4142 PDAC cases for genomic and IO biomarkers.

  • •

    The tumor immune microenvironment of KRAS mt PDAC showed immune-cold features.

  • •

    Expression of IO biomarkers varies between KRAS alterations.

  • •

    IC infiltration patterns of PDAC indicate an immunosuppressive action.

  • •

    Immunotherapeutic strategies in PDAC need to be designed accordingly.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers, with a 5-year overall survival rate of 11%.1 The vast majority of cases of PDAC (90% or higher)2 have been reported to be driven by oncogenic KRAS mutations (mt) that alter the growth pattern of its primary cancer cells and the composition of its tumor microenvironment (TME). Once thought ‘undruggable,’ KRAS mt are now considered as potential therapeutic targets considering approval by the US Food and Drug Administration (FDA) and European Medicines Agency of KRASG12C mutant-targeting designer agents (sotorasib, adagrasib).3 In particular, sotorasib demonstrated single-agent therapeutic signal in patients with PDAC driven by KRASG12C mutations in a phase I-II clinical trial.4 Recent trials reported clinical therapeutic activity of adagrasib in diverse solid tumor populations, including gastrointestinal cancers.4 Additional KRAS inhibitors are currently in the clinical pipeline as a result of ongoing research efforts over the past decade.5

Compounds targeting other KRAS isoforms are in preclinical development. MRTX1133, which specifically targets KRASG12D, has shown high efficacy and safety in KRAS-mutant-driven tumors.6 A clinical research question of strong interest and relevance is whether or not the addition of other cytotoxic, targeted, or immune-targeting drugs can enhance and ideally lead to synergistic anticancer activity when combined with KRAS inhibitors. Emerging preclinical data provide support for potential synergy.7,8 PDAC is high on the list of priority cancers to target considering the high prevalence of KRAS mutations in this cancer as well as the lack of current success in treating this tumor with targeted therapies intended to improve response rates and clinical outcomes.9 The examination of targetable TME changes in the setting of KRASG12D PDAC could grant insight into novel therapeutic strategies that improve the outcome of this PDAC subset.10

Herein, we compare pertinent common and distinguishing features of the KRAS mt versus wild-type (wt) PDAC TME. In particular, we focus on association of KRAS mutations with prevalence of immune cells (ICs) and immuno-oncologic (IO) biomarkers. Historically, immune checkpoint inhibitors have been largely unsuccessful in PDAC aside from the 1% with MSI-H (microsatellite instability-high) status.11 Thus, there is a strong impetus to find PDAC patient subsets that may benefit from immunotherapy. Using a large database of extensively sequenced tumors, we discern differences between KRAS mt versus KRAS wt in PDAC. Aside from the KRASG12D isoform, we find more significant TME differences in PDAC KRAS mt tumors than in colorectal cancer (CRC) KRAS mt tumors.

Materials and methods

Samples

A total of 5072 PDAC and 7463 CRC tumors were submitted to Caris Life Sciences (Phoenix, AZ, demographics in Table 1). The cases were selected based on identification of tumors that had undergone genomic profiling and that had RAS results with expression data available to analyze the TME.

Table 1.

Demographics of PDAC and CRC patients sorted by KRAS-mutant status

Male Female Total (%) Median age, years
PDAC
KRAS mt 2205 1937 4142 (81.7%) 68
KRAS wt 506 424 930 (18.3%) 67
CRC
KRAS mt 1969 1758 3727 (49.9%) 61
KRAS wt 2134 1602 3736 (50.1%) 62
Open in a new tab

CRC, colorectal cancer; mt, mutant; PDAC, pancreatic ductal adenocarcinoma; wt, wild-type.

This study was conducted in accordance with guidelines of the Declaration of Helsinki, Belmont Report, and US Common Rule. In keeping with 45 CFR 46.101(b)(4), this study was carried out utilizing retrospective, de-identified clinical data. Therefore, this study is considered institutional review board exempt and thus patient consent was not indicated.

Next-generation sequencing

Next-generation sequencing (NGS) was carried out on genomic DNA isolated from formalin-fixed paraffin-embedded (FFPE) tumor samples using the NextSeq or NovaSeq 6000 platforms (Illumina, Inc., San Diego, CA). Matched normal tissue was not sequenced. For NextSeq-sequenced tumors, a custom-designed SureSelect XT assay was used to enrich 592 whole-gene targets (Agilent Technologies, Santa Clara, CA). For NovaSeq-sequenced tumors, a hybrid pull-down panel of baits designed to enrich for more than 700 clinically relevant genes at high coverage and high read-depth was used, along with another panel designed to enrich for an additional >20 000 genes at lower depth. All variants were detected with >99% confidence based on allele frequency and amplicon coverage, with an average sequencing depth of coverage of >500 and an analytic sensitivity of 5%. Before molecular testing, tumor enrichment was achieved by harvesting targeted tissue using manual microdissection techniques.

Tumor mutational burden

Tumor mutational burden (TMB) was measured by counting all non-synonymous missense, nonsense, in-frame insertion/deletion, and frameshift mutations found per tumor that had not been previously described as germline alterations in dbSNP151, Genome Aggregation Database (gnomAD) databases or benign variants identified by Caris geneticists. A cut-off point of ≥10 mutations per megabase (mt/Mb) was used based on the KEYNOTE-158 trial, which showed that patients with a TMB of ≥10 mt/Mb across several tumor types had higher response rates than patients with a TMB of <10 mt/Mb.12 Caris Life Sciences is a participant in the Friends of Cancer Research TMB Harmonization Project.13

Microsatellite instability

A combination of multiple test platforms was used to determine MSI-H/ deficiency in mismatch repair (dMMR) status of the tumors profiled, including fragment analysis (FA, Promega, Madison, WI), immunohistochemistry (IHC) [MLH1, M1 antibody; MSH2, G2191129 antibody; MSH6, 44 antibody; and PMS2, EPR3947 antibody (Ventana Medical Systems, Inc., Tucson, AZ)], and NGS (>2800 target microsatellite loci were examined and compared to the reference genome from the University of California, Santa Cruz Genome Browser database). The three platforms generated highly concordant results as previously reported and in the rare cases of discordant results, the MSI-H or MMR status of the tumor was determined in the order of IHC, FA, and NGS.14

mRNA expression using whole-transcriptome sequencing

Gene expression data were evaluated on mRNA isolated from an FFPE tumor sample using the Illumina NovaSeq platform (Illumina, Inc., San Diego, CA) and Agilent SureSelect Human All Exon V7 bait panel (Agilent Technologies, Santa Clara, CA); transcript per million was reported. Gene fusions were detected using the Illumina NovaSeq platform. Additionally, the IC fraction was estimated by the QuanTIseq method.15 Cancer-associated fibroblasts (CAFs) abundance was calculated using Microenvironment Cell Population-counter.16

Statistical analysis

Cohorts were defined by having a pathogenic/presumed pathogenic mutation in KRAS, NRAS, or HRAS or being RAS wt. Comparative analysis of molecular alterations in the cohorts were analyzed using chi-square or Fisher’s exact tests. TMB distribution as well as TME cell fractions were analyzed among cohorts using non-parametric Kruskal–Wallis testing. A P value of <0.05 was considered a trending difference; P values were further corrected for multiple comparison using the Benjamini–Hochberg method to avoid type I error and an adjusted P value (q value) of <0.05 was considered a significant difference.

Results

Prevalence of KRAS mutations and tumor agnostic biomarkers in PDAC

We examined the mutational prevalence, distribution, and differences of KRAS mutations in 4142 PDAC cases. KRAS mt was detected in 81.7% of these PDAC cases. For comparison, the prevalence of KRAS mt was seen in 49.9% of 3727 CRC tumors (Figure 1).

Figure 1.

Figure 1

Open in a new tab

KRAS mutational distribution in PDAC as compared to CRC. CRC, colorectal cancer; mt, mutant; PDAC, pancreatic ductal adenocarcinoma.

Among all point mutation isoforms of KRAS, the most prevalent in both CRC (32%) and PDAC (43%) was KRASG12D, followed by KRASG12V (19% in CRC, 31% in PDAC). The overall prevalence of KRASG13 mutations was higher in CRC (17%) than PDAC (1%) (P < 0.0001). The now actionable target KRASG12C was seen in 7% of CRC cases and in only 2% of PDAC (P < 0.0001). KRASG12A and KRASG12S mutations were more common in CRC compared to PDAC (5% versus 0.4%; 4% versus 0.05%, respectively). Overall, KRASG12 mutations were more common in PDAC as compared to CRC (15% versus 1%) but non-KRASG12 mutations were more common in CRC (9% versus 0.2%). In contrast, KRASQ61 mutations were more common in PDAC (8%) than in CRC (5%). Moreover, we have analyzed the cohort for other isoforms of RAS mutations, but overall cases of mutant HRAS (n = 2) and NRAS (n = 10) were rare and not amenable for further analysis. We further examined prevalence and distribution of individual mt KRAS isoforms in relation to wt in primary tumors as compared to hepatic and non-hepatic metastases in our cohort. Notable findings include a preponderance of all identified KRAS mutations (aside from the KRASother category) in liver metastases as compared to non-liver metastatic tumors, and for all KRAS alterations in distant metastatic sites as compared to their primary tumor counterparts (with the exception of KRASG12S, with 50% in each).

We next focused more specifically on analysis of the PDAC cohort for tumor-agnostic FDA-approved biomarkers, identifying molecular alterations associated with wt KRAS versus cases harboring isoforms of mt KRAS. Firstly, we examined actionable biomarkers including NTRK and RET, for which tumor-agnostic targeted drug approvals are already in place.17,18 Only RET-fusion frequencies showed a small but significant difference between KRAS mt (0% for KRASG12D and KRASG12V) versus KRAS wt (0.7%) (data not shown, P < 0.01).

Next, IO biomarkers tested routinely (TMB, MSI-H/dMMR) or used in other forms of gastrointestinal cancers [programmed death-ligand 1 (PD-L1) IHC], were analyzed (Figure 2). There was significantly higher PD-L1 expression in tumors with KRASG12D (19%), KRASG12C (27%), KRASQ61 (19%), and KRASG13 (33%) mutations as compared to KRAS wt (q < 0.05).

Figure 2.

Figure 2

Open in a new tab

Identification of IO markers in tumors with KRAS mt versus KRAS wt in PDAC tumors, including the range of TMB values in each identified KRAS mt isoform. dMMR, deficiency in mismatch repair; IHC, immunohistochemistry; IO, immuno-oncologic; MSI-H, microsatellite instability-high; mt, mutant; PDAC, pancreatic ductal adenocarcinoma; PD-L1, programmed death-ligand 1; TMB-H, tumor mutational burden-high; wt, wild-type.

In June 2020, the US FDA provided tumor-agnostic approval for use of single-agent immune checkpoint inhibition using pembrolizumab for adult and pediatric patients with chemorefractory solid tumors harboring TMB ≥10 mt/MB.19 Thus we explored the frequency of TMB-high (TMB-H) in our cohort initially using this clinically applicable parameter. We found that overall the rate of TMB-H cases was significantly higher in KRAS wt cases than in KRAS mt cases (4.3% versus 1.3%, q < 0.05). Compared to TMB-H cases, MSI-H/dMMR was more rare overall but still significantly higher in KRAS wt cases compared to KRAS mt cases(1.7% versus 0.9%, q < 0.05). PD-L1 IHC expression was significantly higher in the KRAS mt subcohort (16.7% in KRAS mt versus 12.0% in KRAS wt, q < 0.05). TMB-H had a lower prevalence for KRASG12 isoforms but a significantly higher prevalence in KRASG13 and KRASQ61 mutants (33% versus 4%, q < 0.05). The small MSI-H/dMMR sample size precluded analyzing these trends by individual KRAS mt variant subtype. When assessing for any value of TMB, the median TMB values were highest for KRASG12S cases (7 mt/Mb), and lowest for KRASG12D, KRASG12other, KRASG12V, KRASQ61 as well as KRAS wt cases (Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2024.100042).

The 10 mt/Mb cut-off was based at least in part on results derived from the KEYNOTE-158 study20. However, it is also clear that this cut-off value may not be clinically or biologically meaningful across the spectrum of all solid tumor types.21, 22, 23, 24 In that context, we also thus explored TMB-H values in our PDAC cohort using TMB-H cut-offs of 20 and 50 mt/Mb (Figure 2; Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2024.100042). The widest range of TMB values was seen in cases of KRASG12C, KRASG12D, and KRASother mt (boxplot in Figure 2). As we observed when using a standard cut-off of 10 mt/Mb, KRASG13 isoform-harboring PDAC tumors carried the highest rate of TMB-H (5.0%) for TMB ≥20. Overall, we only identified a single case of PDAC that had TMB ≥50 mt/Mb (Supplementary Table S1, available at https://doi.org/10.1016/j.esmogo.2024.100042).

Analysis of TME by KRAS status

We next examined PDAC IC infiltration and variation of IC composition within the TME in the presence or absence of KRAS alterations (Figure 3).

Figure 3.

Figure 3

Open in a new tab

TME landscape of mutated versus wild-type KRAS PDAC. mt, mutant; NK, natural killer; PDAC, pancreatic ductal adenocarcinoma; TME, tumor microenvironment; wt, wild-type.

KRAS mt tumors had a small but significantly higher abundance of M1 [median cell fraction (MCF): 5.8% versus 4.8%, P < 0.0001] but fewer M2 macrophages (MCF: 3.3% versus 3.8%, P < 0.01), natural killer (NK) cells (MCF: 2.6% versus 2.8%, P < 0.0001), CD4+ (% non-zero cell fraction: 17% versus 24%, P < 0.001), and CD8+ (MCF: 0% versus 0.2%, P < 0.0001) T cells, T-regs (MCF: 1.9% versus 2.2%, P < 0.001), monocytes (% non-zero cell fraction: 0.5% versus 3%, P < 0.0001), as well as myeloid dendritic cells (MCF: 0% versus 0.3%, P < 0.0001). These trends were similar to what was found in KRAS mt CRC cases compared to KRAS wt. There was a non-significant trend toward higher neutrophil infiltration in PDAC KRAS mt cases that was different than the significantly higher prevalence found in KRAS mt CRC cases compared to KRAS wt. Additionally, CAFs were more abundant in KRAS mt PDAC compared to KRAS wt cases (P < 0.001). No significant difference in the prevalence of B cells in KRAS mt cases compared to KRAS wt tumors was found, unlike the decreased prevalence found in the analogous CRC cohort.

Delving into the IC-infiltrative landscape, we found that different KRAS mutations confer different TME changes (Figure 4, Table 2; Supplementary Table S2, available at https://doi.org/10.1016/j.esmogo.2024.100042). For example, while overall there were no significant differences in the B cells and neutrophil prevalence in aggregate KRAS mt compared to KRAS wt, the number of B cells was decreased (q = 0.016) and the number of neutrophils was increased (q = 0.005) in the subset of KRASG12D mutants. In contrast, the significance for both cell types was reversed for KRASG13mutants (q = 0.019 and q = 0.013, respectively). Otherwise, KRASG12D mutants shared similar significant differences to the aggregate KRAS mt analysis, with significantly decreased M2 macrophages, monocytes, NK cells, CD4+, and CD8+ T cells, T-regs, and myeloid dendritic cells. It is important to note that many of the significant differences in the aggregate KRAS mt versus KRAS wt analysis were driven by only a few mutant subtypes. For example, the three mutant KRAS subtypes KRASG12V, KRASQ61, and KRASG12(other) were the only subtypes to have significantly increased M1 macrophages compared to KRAS wt.

Figure 4.

Figure 4

Open in a new tab

TME in KRAS mutational subtypes compared to KRAS wt in PDAC. NK, natural killer; PDAC, pancreatic ductal adenocarcinoma; TME, tumor microenvironment; wt, wild-type.

Table 2.

Comparison of immuno-oncologic biomarkers and TME components in KRAS mutational subtypes compared to KRAS wt in PDAC

KRAS categories N dMMR/MSI TMB PD-L1 B cell M1 M2 Monocyte Neutrophil NK cell CD4+ CD8+ T-regs Myeloid dendritic
KRAS mt 4142 Lowa Lowb Highc Highc Lowa Lowc High (trend) a Lowc Lowc Lowa Lowc Lowa
KRAS-G12 Other 637 Lowb Highb Lowc Lowc Lowc Lowc Lowc Lowc
KRAS-G12A 16
KRAS-G12C 74 Highc
KRAS-G12D 1766 Lowa Lowb Higha Lowa Lowc Lowc Higha Lowc Lowc Lowa Lowc Lowa
KRAS-G12S 2
KRAS-G12V 1293 Lowb Highc Lowc Lowc Lowc Lowc Lowc Lowc Lowc
KRAS-G13 Any 21 Higha Highb Lowc Lowb Lowc Lowa Lowc
KRAS-other 9 Highc Higha
KRAS-Q61 Any 324 Lowc Lowb Highc Highc Lowc Lowc Lowc
Open in a new tab

CRC, colorectal cancer; dMMR, deficiency in mismatch repair; MSI, microsatellite instability; MSS, microsatellite stability; mt, mutant; NK, natural killer; PDAC, pancreatic ductal adenocarcinoma; PD-L1, programmed death-ligand 1; TMB, tumor mutational burden.

a

Similar to MSS-CRC = also significant in CRC and in the same direction.

b

Significant and another direction in MSS-CRC.

c

Different in PDAC = not significant in MSS-CRC.

We detected numerous instances of significant differences in IO marker expression and immune infiltration between wt and mt PDAC tumors, as noted in Table 2.

Analysis of emerging immunologic biomarkers in PDAC specimens

We examined additional actionable biomarkers for immunotherapeutic strategies based on the presence of RAS mutations. The mean mRNA expression of CD80, CTLA-4, HAVCR2, IFNG, LAG3, and PDCD1LG2 were all significantly decreased in KRAS mt PDAC cases, whereas CD86 was increased in KRAS mt samples (Figure 5). Interestingly, the relative median mRNA expression levels of CD80 and PDCD1LG2 did not always correlate with the trends in their mean expression, suggesting outlier effects (Figure 5).

Figure 5.

Figure 5

Open in a new tab

IO markers for KRAS mutations in PDAC. IO, immuno-oncologic; mt, mutant; N.S., not significant; PDAC, pancreatic ductal adenocarcinoma; wt, wild-type.

CD274, IDO1, and PDCD1 did not have significant expression changes in KRAS mt versus KRAS wt PDAC. These differences were largely driven by KRASG12D mutants, aside from CTLA-4, whose decreased mRNA expression was found to be borderline significant (Figure 6). When stratifying by other KRAS isoforms, we found KRASG12V and KRASG12(other) had the next highest number of dysregulated IO biomarkers (four), followed by KRASQ61 (three) (Figure 6).

Figure 6.

Figure 6

Open in a new tab

IO markers in KRAS mutational subtypes. IO, immuno-oncologic; N.S., not significant.

Discussion

In this study, we examined a large cohort of 4142 PDAC and 3727 CRC cases harboring a KRAS mutation by next-generation DNA sequencing, IHC, and whole-transcriptome RNA sequencing. PDAC KRAS mutation rates were consistent with the literature, as was the very low prevalence of MSI-H/dMMR. While the prevalence of these alterations is important, different KRAS mutations confer divergent downstream signaling effects similar to the differences seen in the TME shown herein and as a result, have different responses to treatment and prognoses.25 For example, KRASG12C mutated CRC is generally more chemoresistant and associated with worse overall survival compared to CRC driven by other KRAS isoforms.26 Comparative analysis of the KRAS mt landscape between CRC and PDAC confirmed a lower prevalence of MSI-H/dMMR than KRAS wt PDAC. Further, PD-L1 IHC expression was significantly lower in KRAS wt compared to KRASG12D and KRASG13, similar to observations in CRC. In contrast to CRC, TMB rates were lower in RAS mt than RAS wt PDAC tumors. The IC environment of KRAS mt PDAC showed significantly higher infiltration with M1 macrophages and CAF. Conversely, there were lower M2 macrophages, CD4+ and CD8+ T cells, T-regs, NK, myeloid dendritic, and endothelial cells in KRAS mt compared to the KRAS wt PDAC population. In CRC, a similar pattern was observed but these TME changes are more pronounced in PDAC.

In this study, the classic biomarkers of response to immunotherapy, dMMR/MSI and TMB, were higher in the KRAS wt group, while PD-L1 scores were higher in mt KRAS cases. Increased expression of the PD-L1 surface marker, and thus increased activation of the PD-L1/programmed cell death protein 1 axis in mt KRAS cases, is a finding that is consistent with the generally immunosuppressive nature of PDAC as well as worse patient prognosis.27 Likewise, decreased prevalence of dMMR/MSI and TMB in these mt cases also support the notion that PDAC presents an immune desert in which current strategies for immunotherapy fail to produce meaningful clinical responses in this cancer type. Thus it is crucial to look beyond these standard IO biomarkers in order to define ones that may have more clinical impact in PDAC treatment. The receptors CD80 and CD86 on antigen-presenting cells normally bind CD28 in an immunostimulatory fashion, especially when CTLA-4 expression is decreased.28 While CD86 expression was increased in KRAS mt PDAC, both CD80 and immunostimulatory cytokine IFNG expression were decreased, making the significance of these observations unclear for the susceptibility of KRAS mt PDAC to IO regimens. Further exploration of these trends are warranted as only mRNA expression data were used to analyze these TME features. Furthermore, the immune-regulatory biomarkers CTLA-4, HAVCR2, PDCD1LG2, and LAG3, which, when overexpressed, normally serve an immunosuppressive role on T cells, were all decreased in KRAS mt PDAC.29, 30, 31, 32 Overall differences were most significant in cases harboring KRASG12D, followed by KRASG12V and KRASQ61 variants. This diverges from other KRAS mt-driven tumor types generally resistant to immunotherapy. For example, CTLA-4 expression is increased in KRAS mt CRC.33 Increased LAG3 expression was seen in an ovarian cancer model driven by KRASG12V.34 However, these different tumor types have other biological features and drivers that may confound these results.

Considering the prevalence of KRAS mt PDAC tumors, our findings suggest that different TME correlate with different KRAS point mutations. As such, future PDAC treatment may need to be personalized depending on which KRAS mutation is driving tumor growth and survival. Traditionally, direct inhibition of KRAS has been challenging. This paradigm is now changing with the recent FDA approval of the KRASG12C inhibitor sotorasib to treat non-small-cell lung cancers (NSCLC) with other inhibitors such as adagrasib (KRASG12C) and JDQ443 (KRASG12C) in clinical trials.3,35 However, no inhibitor has been approved to treat PDAC KRASG12C tumors specifically given its rarity, but trials are underway.4 Furthermore, resistance to these selective KRASG12C inhibitors is emerging via mutations in other proteins of the MAPK pathway, new point mutations in the KRAS binding pocket itself, and through nongenetic mechanisms.3 While our understanding of these resistance mechanisms is incomplete, one clinical nongenetic resistance mechanism activated the YAP1 pathway and transforming growth factor-β-driven TME changes including epithelial-mesenchymal transition and angiogenesis.36 Similarly, resistance to KRASG12C inhibitors has been observed in CRC driven by this particular KRAS mutation, underscoring the importance of characterizing the TME of each KRAS-mutant isoform, as treatment may need to be personalized depending on the driving mutation and resistance mechanisms.37

In PDAC, the KRASG12D mutation is much more common and is a large contributor to the high disease mortality. By teasing out the unique TME features of this mutation, therapeutic regimens that target both KRASG12D and these TME characteristics may be a more effective strategy to personalize treatment for this deadly disease. Our results begin to point the way toward more isoform-specific profiles and their implications for prognosis and targetability. For example, the three mutant KRAS subtypes KRASG12V, KRASQ61, and KRASG12(other) were the only subtypes to have significantly increased M1 macrophages compared to KRAS wt. The balance between M1 and M2 macrophages leaned more heavily toward the latter in the other predominant isoforms of KRAS, and these differences in macrophage formation and infiltration may provide an important insight into immune balance that decreases susceptibility to IO treatments. In pancreatic cancer, high levels of M1 macrophages can indicate a favorable prognosis with better overall survival and response to chemotherapy and high levels of M2 macrophages have been associated with poor prognosis and resistance to chemotherapy.38 Therefore, the balance between M1 and M2 macrophages in the pancreatic cancer microenvironment is crucial for determining the patient's prognosis and response to treatment. The imbalance of M1 and M2 macrophages is especially timely in light of emerging data of the role of tumor-associated macrophages (TAMs) in PDAC, as TAMs tend toward M2 phenotypes.38

Historically, PDAC has been characterized as an ‘immune desert,’ as multiple clinical trials of various IO have shown no benefit.39 However, these trials did not utilize molecular profiles for patient stratification or interpretation of results in the context of aggregated components of the TME. Considering the differences in the TME seen with various KRAS mutations in PDAC, it is conceivable that a subset of PDAC tumors may respond to IO, especially in combination with therapies that target the particular KRAS mutation. Further exploration is required to see if this feature is relatively unique to PDAC, as other KRAS mt tumor types have shown opposite TME trends suggestive of IO resistance.33,34

In this era that combines advances in immunotherapy and targeted therapies in many cancers (NSCLC providing a prototype), there is especially strong interest in targeting mt KRAS in PDAC due to its extremely high prevalence. A recently reported clinical trial using single-agent inhibition of KRASG12C in PDAC patients with this isoform led to a 21% reported objective response.4 How best to add immunotherapy effectively to this targeted strategy remains speculative but of course a strong area of interest. There are preclinical data supporting targeting KRASG12D, based on efficacy in a genetically engineered mouse model; the reported results indicated that single-agent blockade led to significant shrinkage of tumors in many mice tested, and further modulation of the TME induced changes potentially conducive to immune checkpoint inhibition.40 Preclinical approaches using this strategy are ongoing and may inform rational clinical trial design for eligible human patients within the next few years, as is occurring in current trials for other cancer types including NSCLC. As more IO therapies emerge targeting various pathways of the immunological response, personalized clinical trials that consider the KRAS-mutant status and other molecular features may be warranted. As more selective KRAS mt targeting agents advance in the clinical pipeline, combinations with immunotherapy may help improve overall survival of patients with this hard-to-treat cancer.

There are several limitations to this study that make some blanket conclusions prohibitive. In terms of accuracy of genomic/transcriptomic analysis, inter-tumor heterogeneity based on anatomic location and corresponding TME (e.g. primary tumor versus lymph node versus distant site) may account for some differences in findings but this information was not available from our dataset. Likewise, other factors including whether tumors were profiled (at the time of initial diagnosis versus recurrence), before or following disease progression (including treatment naïve versus resistant/at progression categories), and the type of samples (surgical specimen versus needle biopsy) may lead to variability in such studies. Beyond analysis of genomic drivers of PDAC in association with the inflammatory microenvironment, there are numerous other factors at play that have direct implication for tumor invasiveness and metastatic potential. In addition to the ever-dynamic inflammatory environment, the role of intratumoral microbiota is increasingly being identified as a wild card factor that both alters surrounding inflammation as well as expression of molecular driver and passenger alterations. Outside of the sphere of the tumors themselves, with increasing recognition of host–tumor interactions, factors such as obesity, sarcopenia, presence and extent of glycemic control in context of diabetes, and intestinal microbiota, and many other factors, have to be considered for their potential effects. All of these factors are worth exploration separately or in tandem in future large-scale studies, and will likely become more feasible with maturation of sensitivity and specificity of detection technology over time.

In summary, these results demonstrate that the TME of KRAS mt PDAC as a whole presents with immune-cold features. As the KRASG12D variant becomes druggable, an additional 35% of PDAC patients and 15% of CRC patients may potentially benefit through the continued development of isoform-specific RAS inhibitors. Our results provide evidence that IC composition of the PDAC TME leans toward immunosuppression, in line with results of failed trials using IO to target this tumor type. Tailored immunotherapeutic strategies would have to overcome these barriers in KRAS mt PDAC, possibly in combination with molecularly targeted treatment strategies.

Acknowledgements

The authors would like to thank all the participating centers of the Caris Precision Oncology Alliance and their support staffs for contributing to this research project.

Funding

Primary funding for this work was provided by Caris Life Sciences. EBF was supported in this work by NIH training [grant number T32 GM008244]. We are especially grateful to the Love Like Laurie Legacy and to the family and friends of Gayle Huntington for fundraising for this and other pancreatic cancer research in the Lou Lab.

Disclosure

YB, JX, PW, and CN are employees of Caris. AnwS reports research grants from AstraZeneca, Bristol Myers Squibb, Merck, Clovis, Exelixis, Actuate therapeutics, Incyte Corporation, Daiichi-Sankyo, Five prime therapeutics, Amgen, Innovent biologics, Dragonfly therapeutics, KAHR medical, and Biontech; and advisory board fees from AstraZeneca, Bristol Myers Squibb, Exelixis, Pfizer, and Daiichi-Sankyo. HJL reports advisory board/consulting fees from Merck, Merck KG, Bayer, Roche, G1 Therapeutics, Jazz Therapeutics, Fulgent, and 3T Bioscience. AndS is on the speaker’s panel for Caris Life Sciences. EL reports honorarium and travel expenses for a research talk at GlaxoSmithKline in 2016; honoraria and travel expenses for laboratory-based research talks, and equipment for laboratory-based research, Novocure, LLC, 2018-present; honorarium for panel discussion organized by Antidote Education for a CME module on diagnostics and treatment of HER2+ gastric and colorectal cancers, funded by Daiichi-Sankyo, 2021 (honorarium donated to laboratory); consultant, Nomocan Pharmaceuticals (unpaid); Scientific Advisory Board Member, Minnetronix, LLC, 2018-2020 (unpaid); consultant and speaker honorarium, Boston Scientific US, 2019; Institutional Principal Investigator for clinical trials sponsored by Celgene, Novocure, Intima Biosciences, and the National Cancer Institute, and University of Minnesota membership in the Caris Life Sciences Precision Oncology Alliance (unpaid). EL also reports research grants from the American Association for Cancer Research (AACR-Novocure Tumor-Treating Fields Research Award, Grant Number 19-60-62-LOU), the American Cancer Society, and the Minnesota Ovarian Cancer Alliance. All other authors have declared no conflicts of interest.

Supplementary data

Supplementary Table 1
mmc1.docx (1.2MB, docx)
Supplementary Table 2
mmc2.docx (13.4KB, docx)

References

  • 1.Siegel R.L., Miller K.D., Fuchs H.E., Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72(1):7–33. doi: 10.3322/caac.21708. [DOI] [PubMed] [Google Scholar]
  • 2.Philip P.A., Azar I., Xiu J., et al. Molecular characterization of KRAS wild-type tumors in patients with pancreatic adenocarcinoma. Clin Cancer Res. 2022;28(12):2704–2714. doi: 10.1158/1078-0432.CCR-21-3581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Nakajima E.C., Drezner N., Li X., et al. FDA approval summary: sotorasib for KRAS G12C-mutated metastatic NSCLC. Clin Cancer Res. 2022;28(8):1482–1486. doi: 10.1158/1078-0432.CCR-21-3074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Strickler J.H., Satake H., George T.J., et al. Sotorasib in KRAS p.G12C-mutated advanced pancreatic cancer. N Engl J Med. 2023;388(1):33–43. doi: 10.1056/NEJMoa2208470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Parikh K., Banna G., Liu S.V., et al. Drugging KRAS: current perspectives and state-of-art review. J Hematol Oncol. 2022;15(1):152. doi: 10.1186/s13045-022-01375-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hallin J., Bowcut V., Calinisan A., et al. Anti-tumor efficacy of a potent and selective non-covalent KRAS(G12D) inhibitor. Nat Med. 2022;28(10):2171–2182. doi: 10.1038/s41591-022-02007-7. [DOI] [PubMed] [Google Scholar]
  • 7.Mugarza E., van Maldegem F., Boumelha J., et al. Therapeutic KRAS(G12C) inhibition drives effective interferon-mediated antitumor immunity in immunogenic lung cancers. Sci Adv. 2022;8(29) doi: 10.1126/sciadv.abm8780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Canon J., Rex K., Saiki A.Y., et al. The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature. 2019;575(7781):217–223. doi: 10.1038/s41586-019-1694-1. [DOI] [PubMed] [Google Scholar]
  • 9.Singhi A.D., George B., Greenbowe J.R., et al. Real-time targeted genome profile analysis of pancreatic ductal adenocarcinomas identifies genetic alterations that might be targeted with existing drugs or used as biomarkers. Gastroenterology. 2019;156(8):2242–2253.e2244. doi: 10.1053/j.gastro.2019.02.037. [DOI] [PubMed] [Google Scholar]
  • 10.Qian Z.R., Rubinson D.A., Nowak J.A., et al. Association of alterations in main driver genes with outcomes of patients with resected pancreatic ductal adenocarcinoma. JAMA Oncol. 2018;4(3) doi: 10.1001/jamaoncol.2017.3420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Le D.T., Durham J.N., Smith K.N., et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357(6349):409–413. doi: 10.1126/science.aan6733. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Marabelle A., Fakih M., Lopez J., et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet Oncol. 2020;21(10):1353–1365. doi: 10.1016/S1470-2045(20)30445-9. [DOI] [PubMed] [Google Scholar]
  • 13.Vega D.M., Yee L.M., McShane L.M., et al. Aligning tumor mutational burden (TMB) quantification across diagnostic platforms: phase II of the Friends of Cancer Research TMB Harmonization Project. Ann Oncol. 2021;32(12):1626–1636. doi: 10.1016/j.annonc.2021.09.016. [DOI] [PubMed] [Google Scholar]
  • 14.Salem M.E., Puccini A., Grothey A., et al. Landscape of tumor mutation load, mismatch repair deficiency, and PD-L1 expression in a large patient cohort of gastrointestinal cancers. Mol Cancer Res. 2018;16(5):805–812. doi: 10.1158/1541-7786.MCR-17-0735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Finotello F., Mayer C., Plattner C., et al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med. 2019;11(1):34. doi: 10.1186/s13073-019-0638-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Becht E., Giraldo N.A., Lacroix L., et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol. 2016;17(1):218. doi: 10.1186/s13059-016-1070-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Duke E.S., Bradford D., Marcovitz M., et al. FDA approval summary: selpercatinib for the treatment of advanced RET fusion-positive solid tumors. Clin Cancer Res. 2023;29(18):3573–3578. doi: 10.1158/1078-0432.CCR-23-0459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Marcus L., Donoghue M., Aungst S., et al. FDA approval summary: entrectinib for the treatment of NTRK gene fusion solid tumors. Clin Cancer Res. 2021;27(4):928–932. doi: 10.1158/1078-0432.CCR-20-2771. [DOI] [PubMed] [Google Scholar]
  • 19.Marcus L., Fashoyin-Aje L.A., Donoghue M., et al. FDA approval summary: pembrolizumab for the treatment of tumor mutational burden-high solid tumors. Clin Cancer Res. 2021;27(17):4685–4689. doi: 10.1158/1078-0432.CCR-21-0327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Marabelle A., Le D.T., Ascierto P.A., et al. Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer: results from the phase II KEYNOTE-158 study. J Clin Oncol. 2020;38(1):1–10. doi: 10.1200/JCO.19.02105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Swat W., Kumar D., Boyle T.A., et al. Distributional range of tumor mutational burden (TMB) across multiple cancer types emphasizing the importance of different cutoff values in different tumor types for clinical interpretation: a consensus statement from the SEQUOIA consortium. J Clin Oncol. 2023;41(suppl 16) [Google Scholar]
  • 22.Riviere P., Goodman A.M., Okamura R., et al. High tumor mutational burden correlates with longer survival in immunotherapy-naive patients with diverse cancers. Mol Cancer Ther. 2020;19(10):2139–2145. doi: 10.1158/1535-7163.MCT-20-0161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Buttner R., Longshore J.W., Lopez-Rios F., et al. Implementing TMB measurement in clinical practice: considerations on assay requirements. ESMO Open. 2019;4(1) doi: 10.1136/esmoopen-2018-000442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Lawlor R.T., Mattiolo P., Mafficini A., et al. Tumor mutational burden as a potential biomarker for immunotherapy in pancreatic cancer: systematic review and still-open questions. Cancers (Basel) 2021;13(13):3119. doi: 10.3390/cancers13133119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Modest D.P., Brodowicz T., Stintzing S., et al. Impact of the specific mutation in KRAS codon 12 mutated tumors on treatment efficacy in patients with metastatic colorectal cancer receiving cetuximab-based first-line therapy: a pooled analysis of three trials. Oncology. 2012;83(5):241–247. doi: 10.1159/000339534. [DOI] [PubMed] [Google Scholar]
  • 26.Modest D.P., Ricard I., Heinemann V., et al. Outcome according to KRAS-, NRAS- and BRAF-mutation as well as KRAS mutation variants: pooled analysis of five randomized trials in metastatic colorectal cancer by the AIO colorectal cancer study group. Ann Oncol. 2016;27(9):1746–1753. doi: 10.1093/annonc/mdw261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Mucileanu A., Chira R., Mircea P.A. PD-1/PD-L1 expression in pancreatic cancer and its implication in novel therapies. Med Pharm Rep. 2021;94(4):402–410. doi: 10.15386/mpr-2116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Linsley P.S., Greene J.L., Brady W., et al. Human B7-1 (CD80) and B7-2 (CD86) bind with similar avidities but distinct kinetics to CD28 and CTLA-4 receptors. Immunity. 1994;1(9):793–801. doi: 10.1016/s1074-7613(94)80021-9. [DOI] [PubMed] [Google Scholar]
  • 29.Rudd C.E., Taylor A., Schneider H. CD28 and CTLA-4 coreceptor expression and signal transduction. Immunol Rev. 2009;229(1):12–26. doi: 10.1111/j.1600-065X.2009.00770.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Huang C.T., Workman C.J., Flies D., et al. Role of LAG-3 in regulatory T cells. Immunity. 2004;21(4):503–513. doi: 10.1016/j.immuni.2004.08.010. [DOI] [PubMed] [Google Scholar]
  • 31.Jin H.T., Anderson A.C., Tan W.G., et al. Cooperation of Tim-3 and PD-1 in CD8 T-cell exhaustion during chronic viral infection. Proc Natl Acad Sci U S A. 2010;107(33):14733–14738. doi: 10.1073/pnas.1009731107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Zhang Y., Chung Y., Bishop C., et al. Regulation of T cell activation and tolerance by PDL2. Proc Natl Acad Sci U S A. 2006;103(31):11695–11700. doi: 10.1073/pnas.0601347103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Tanjak P., Chaiboonchoe A., Suwatthanarak T., et al. The KRAS-mutant consensus molecular subtype 3 reveals an immunosuppressive tumor microenvironment in colorectal cancer. Cancers (Basel) 2023;15(4):1098. doi: 10.3390/cancers15041098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Rodriguez G.M., Galpin K.J.C., Cook D.P., et al. The tumor immune profile of murine ovarian cancer models: an essential tool for ovarian cancer immunotherapy research. Cancer Res Commun. 2022;2(6):417–433. doi: 10.1158/2767-9764.CRC-22-0017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Janne P.A., Riely G.J., Gadgeel S.M., et al. Adagrasib in non-small-cell lung cancer harboring a KRAS(G12C) mutation. N Engl J Med. 2022;387(2):120–131. doi: 10.1056/NEJMoa2204619. [DOI] [PubMed] [Google Scholar]
  • 36.Tsai Y.S., Woodcock M.G., Azam S.H.m., et al. Rapid idiosyncratic mechanisms of clinical resistance to KRAS G12C inhibition. J Clin Invest. 2022;132(4) doi: 10.1172/JCI155523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Yaeger R., Mezzadra R., Sinopoli J., et al. Molecular characterization of acquired resistance to KRASG12C-EGFR inhibition in colorectal cancer. Cancer Discov. 2023;13(1):41–55. doi: 10.1158/2159-8290.CD-22-0405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Yang S., Liu Q., Liao Q. Tumor-associated macrophages in pancreatic ductal adenocarcinoma: origin, polarization, function, and reprogramming. Front Cell Dev Biol. 2020;8 doi: 10.3389/fcell.2020.607209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Hilmi M., Bartholin L., Neuzillet C. Immune therapies in pancreatic ductal adenocarcinoma: where are we now? World J Gastroenterol. 2018;24(20):2137–2151. doi: 10.3748/wjg.v24.i20.2137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Kemp S.B., Cheng N., Markosyan N., et al. Efficacy of a small-molecule inhibitor of KrasG12D in immunocompetent models of pancreatic cancer. Cancer Discov. 2023;13(2):298–311. doi: 10.1158/2159-8290.CD-22-1066. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Table 1
mmc1.docx (1.2MB, docx)
Supplementary Table 2
mmc2.docx (13.4KB, docx)

Articles from ESMO Gastrointestinal Oncology are provided here courtesy of Elsevier

RESOURCES