Skip to main content
NCBI home page
The Oncologist logoLink to The Oncologist
. 2021 Mar 5;26(7):569–578. doi: 10.1002/onco.13717

Circulating Tumor DNA‐Based Testing and Actionable Findings in Patients with Advanced and Metastatic Pancreatic Adenocarcinoma

Gehan Botrus 1, Heidi Kosirorek 1, Mohamad Bassam Sonbol 1, Yael Kusne 1, Pedro Luiz Serrano Uson Junior 1, Mitesh J Borad 1, Daniel H Ahn 1, Pashtoon M Kasi 3, Leylah M Drusbosky 4, Hiba Dada 4, Phani Keerthi Surapaneni 2, Jason Starr 2, Ashton Ritter 2, Jessica McMillan 2, Natasha Wylie 2, Kabir Mody 2,[Link], Tanios S Bekaii‐Saab 1,[Link],✉
PMCID: PMC8265372  PMID: 33555095

Abstract

Purpose

Recent advances in molecular diagnostic technologies allow for the evaluation of solid tumor malignancies through noninvasive blood sampling, including circulating tumor DNA profiling (ctDNA). Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis, often because of late presentation of disease. Diagnosis is often made using endoscopic ultrasound or endoscopic retrograde cholangiopancreatography, which often does not yield enough tissue for next‐generation sequencing. With this study, we sought to characterize the ctDNA genomic alteration landscape in patients with advanced PDAC with a focus on actionable findings.

Materials and Methods

From December 2014 through October 2019, 357 samples collected from 282 patients with PDAC at Mayo Clinic underwent ctDNA testing using a clinically available assay. The majority of samples were tested using the 73‐gene panel which includes somatic genomic targets, including complete or critical exon coverage in 30 and 40 genes, respectively, and in some, amplifications, fusions, and indels. Clinical data and outcome variables were available for 165 patients; with 104 patients at initial presentation.

Results

All patients included in this study had locally advanced or metastatic PDAC. Samples having at least one alteration, when variants of unknown significance (VUS) were excluded, numbered 266 (75%). After excluding VUS, therapeutically relevant alterations were observed in 170 (48%) of the total 357 cohort, including KRAS (G12C), EGFR, ATM, MYC, BRCA, PIK3CA, and BRAF mutations. KRAS, SMAD, CCND2, or TP53 alterations were seen in higher frequency in patients with advanced disease.

Conclusion

Our study is the largest cohort to date that demonstrates the feasibility of ctDNA testing in PDAC. We provide a benchmark landscape upon which the field can continue to grow. Future applications may include use of ctDNA to guide treatment and serial monitoring of ctDNA during disease course to identify novel therapeutic targets for improved prognosis.

Implications for Practice

Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis often due to late presentation of disease. Biopsy tissue sampling is invasive and samples are often inadequate, requiring repeated invasive procedures and delays in treatment. Noninvasive methods to identify PDAC early in its course may improve prognosis in PDAC. Using ctDNA, targetable genes can be identified and used for treatment.

Keywords: ct‐DNA, Advanced, Pancreatic cancer

Short abstract

This article characterizes the ctDNA genomic alteration landscape in patients with advanced pancreatic ductal adenocarcinoma, with a focus on actionable findings.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is the fourth leading cause of cancer deaths in the United States [1], with only 10%–20% of tumors resectable at the time of diagnosis [2]. Given the importance in early detection for improved prognosis in PDAC, the use of highly sensitive diagnostic approaches is necessary. In addition, as the disease progresses, new genetic changes may occur, leading to more aggressive and less treatment‐responsive cells, resulting in secondary treatment resistance [3]. The ability to detect these changes without using invasive techniques can potentially help to identify genetic targets with therapeutic implications [4]. For example, patients with PDAC with homologous recombination deficiency have better overall survival and progression‐free survival (PFS) when treated with platinum‐based chemotherapy [5, 6]. Additionally, in a phase III clinical trial, patients with BRCA1 or BRCA2 germline mutations experienced longer PFS with the use of maintenance olaparib, the poly(adenosine diphosphate‐ribose) inhibitor, when compared with placebo [4].

PDAC tissue samples obtained via endoscopic ultrasound or endoscopic retrograde cholangiopancreatography can be insufficient and often requires repeated invasive procedures. Recent advances in molecular diagnostic technologies allow for the evaluation of solid tumor malignancies through noninvasive blood sampling. Intact circulating tumor cells and circulating DNA from both leukocytes and tumor (circulating tumor DNA [ctDNA]) can now be isolated and analyzed using advanced sequencing methods. ctDNA is DNA released into circulation specifically from tumor cells that undergo metabolic secretion, apoptosis, or necrosis. ctDNA has been shown to carry tumor‐specific genetic or epigenetic alterations, such as point mutations, copy number variations, chromosomal rearrangements, and DNA methylation patterns [7, 8, 9, 10]. The two dominant approaches in this field are polymerase chain reaction (PCR)‐based, also known as digital PCR, and next‐generation sequencing (NGS) approaches. Digital PCR approaches are highly sensitive but can only examine a single or a few mutations of interest at any one time. NGS approaches have the ability to look at a larger number of genes simultaneously. However, these techniques have historically been limited by the need for high sensitivity and specificity and cost associated with sequencing. Capture‐based NGS has the ability to enrich genomic regions of interest by hybridizing target genes and regions to antisense oligonucleotides prior to sequencing; this approach allows for agnostic analysis of large portions of the genome and can identify multiple mutations with increased sensitivity [11]. The evaluation of ctDNA is particularly attractive, as it enables assessment of patient‐specific tumoral genetic and/or epigenetic alterations while allowing for serial monitoring of tumor genomes in a noninvasive, convenient, fast, and accurate manner. Potential applications of ctDNA testing in patients with cancer include (a) early detection of disease; (b) evaluation of tumor heterogeneity; (c) identification of therapeutic targets; (d) real‐time evaluation of treatment response, tumor relapse, and progression; and (e) real‐time assessment of tumor evolution and development of drug resistance. Additionally, data have shown utility in using NGS‐based ctDNA analysis for patient stratification into phase I clinical trials, allowing for earlier initiation of targeted therapy [12, 13]. NGS‐based ctDNA evaluation has been used in multiple cancer types and studies have revealed high rates of specificity and sensitivity [14].

The purpose of the current study was to examine results of ctDNA testing from a large cohort of PDAC patients at Mayo Clinic, acquired in the course of clinical practice, and characterize the mutational landscape of PDAC using this technology.

Materials and Methods

Patients

From December 2014 through October 2019, 282 patients with PDAC underwent ctDNA testing using a clinically available assay (Guardant Health, Inc.; Redwood City, CA). Samples were obtained from patients cared for at Mayo Clinic Cancer Center in Florida and Arizona. A total of 357 samples from the 282 patients were included in the study. The data analysis from this patient cohort was reviewed and approved by the Mayo Clinic institutional review board.

Demographic information and date of blood collection were available for all patients (supplemental online Table 1). Pertinent clinical data was collected from medical records for a subset of patients (n = 104) for whom ctDNA testing was conducted at initial presentation before the start of treatment (Table 1).

Table 1.

Subset of patients with ctDNA analysis performed at diagnosis, prior to therapy

Characteristics n (%)
Gender
Male 52 (50)
Female 52 (50)
Age, median (range) 70 (43–91)
Stage at time of test
Locally advanced 39 (37.5)
Metastatic 65 (62.5)
Chemotherapy
Gemcitabine based 66 (63.0)
5FU‐based regimen 29 (28.0)
No chemotherapy 8 (7.0)
Other 1 (1.0)
Surgery 17 (43.5)
Chemoradiation 17 (43.5)
Open in a new tab

Abbreviation: 5FU, fluorouracil.

Comprehensive Genomic Testing in Plasma

ctDNA was extracted from whole blood collected in 10‐mL Streck tubes. Samples were shipped to a Clinical Laboratory Improvement Act–certified, College of American Pathologists–accredited laboratory (Guardant Health, Inc., Redwood City, California). After double ultracentrifugation, 5–30 ng of ctDNA was isolated from plasma for digital sequencing.

ctDNA fragments, both leukocyte‐ and tumor‐derived, were simultaneously sequenced. The variant allele fraction was calculated as the proportion of ctDNA harboring the variant in a background of wild‐type ctDNA. The analytical sensitivity allowed detection of 1–2 mutant fragments in a 10‐mL blood sample (0.1% limit of detection) with analytic specificity >99.9999%. Twelve (CNAs) were reported as the absolute gene copy number in plasma. Because most cell‐free DNA is leukocyte derived, the gene copy number is generally 2.0. Tumor‐derived DNA shed into the bloodstream raises this value but, because of the relative proportions of tumor‐derived versus leukocyte‐derived ctDNA, it is typically a minor contributor. Gene copy number in plasma is thus a function of both copy number in tissues and the degree to which tumor DNA is shed into circulation. Plasma copy number of 2.5–4.0 is reported as ++ amplification and >4.0 as +++, representing the 50th–90th and > 90th percentile, respectively, of all CNA calls in the Guardant360 database [15, 16, 17].

Bioinformatics analysis of NGS data has been previously described [18, 19]. 150‐base pair paired‐end reads are aligned to the reference genome, hg19. Custom scripts were used to (a) remove spurious variants (“noise”) created by sequencing errors, (b) identify all germline single nucleotide polymorphisms and somatic single nucleotide variants (SNVs), and (c) call somatic SNVs, while removing erroneous variants resulting from sequencing errors, DNA damage, strand bias, and so on.

Over the course of the study, the panel composition expanded from 54 to 68 to 70 to 73 to 74 genes. The initial 54‐gene panel consisted of complete exon coverage or critical exon coverage in 54‐cancer related genes and amplifications in 3 genes. In the 68‐gene panel, 8 genes were retired from the SNV gene set, whereas coverage of gene amplifications expanded from 3 to 16 genes, and the addition of detection of fusions in 4 genes, and insertions or deletion of bases (indels) in 1 gene was made. The 70‐gene panel included all national comprehensive cancer network (NCCN) somatic genomic targets, including complete or critical exon coverage in 30 and 40 genes, respectively; amplifications in 14 genes, fusions in 6 genes, and indels in 3 genes. The 73‐gene panel includes the addition of 5 genes to and removal of 2 genes from the list, and the currently used 74‐gene panel added one gene. The vast majority of samples in this study were tested using the 73‐gene panel [20].

To determine whether an alteration was therapeutically relevant or not, we defined “therapeutically relevant” as any gene alteration with OncoKB level of evidence of 1, 2A, 2B, 3A, 3B, or R1 [21]. OncoKB is a precision oncology knowledge base containing information about the effects and treatment implications of specific cancer gene alterations [21]. OncoKB contains detailed information about specific alterations in 477 cancer genes curated from various sources, including guidelines from the U.S. Food and Drug Administration (FDA), NCCN, or American Society of Clinical Oncology, ClinicalTrials.gov, and the scientific literature. For each variant, the biological effect, prevalence, and prognostic information, as well as treatment implications, are noted. Treatment information is classified into the levels of evidence system that assigns the clinical actionability to individual mutational events. Levels 1 and 2 include those variants that are FDA‐recognized or considered standard care biomarkers predictive of response to FDA‐approved drugs in specific disease settings; level 3 variants are those which are considered predictive of response based on promising clinical data to targeted agents being tested in clinical trials; level 4 includes those variants which are considered predictive of response to targeted agents based on compelling biological, nonclinical evidence.

By these standards, a gene is considered therapeutically relevant if there is supporting evidence that such gene is a driver of tumorigenesis, and wherein actionability of the gene can refer to either sensitivity and/or resistance to a drug(s), regardless of the tumor type each variant has been associated with. Furthermore, sensitivity and/or resistance can be inclusive of all alterations types or specific to alteration classes (i.e., applicable to amplifications, but not mutations). Furthermore, there must be a clinically available agent (including clinical trials) targeting such genes or proteins, and there must be at least preclinical evidence that supports its role in targeting the specific gene or protein.

Based on the aforementioned criteria, the following genes were considered therapeutically relevant: AKT1, ALK, ARAF, ATM, BRAF, BRCA1, BRCA2, CDK4, EGFR, ERBB2, ESR1, FGFR1, FGFR2, IDH1, IDH2, JAK2, KIT, KRAS, MAP2K1, MAP2K2, MET, MTOR, MYC, NRAS, NTRK1, NTRK3, PDGFRA, PIK3CA, RAF1, and TSC1. Variants found in therapeutically relevant genes that did not alter the amino acid sequence (i.e., synonymous mutations) and/or those that lacked any supporting, functional evidence of pathogenicity (variants of undetermined significance; VUS) were excluded.

PDAC NGS data was also collected from cBioportal, a public database of genomic alterations in cancer, to compare the frequencies of alterations identified in ctDNA and tissue samples.

Statistical Analysis

The distribution of each continuous variable was summarized by its mean, SD, and range. The distribution of each categorical variable was summarized in terms of its frequencies and percentages. Continuous variables were compared between groups by Wilcoxon rank sum test, and for categorical data, the comparison was conducted by Fisher exact test. The relationship between gene types with regard to mutation and amplification as well as synonymous and targetable status was evaluated with Spearman's rank correlation and displayed in the correlation matrix, in which the nonsignificant correlations are marked with “blank” in the graph. A subgroup analysis was conducted in 104 patients with baseline ctDNA results available prior to initiation of systemic therapy to determine the association of genetic mutations and disease stage (locally advanced or metastatic). In addition, overall survival (OS) and presence of genetic mutations were evaluated with Cox regression models in a univariate and multivariate fashion. Patients without a death event were considered censored at the date of last known follow‐up. The p values < .05 were considered statistically significant. All computations were carried out in SAS version 9.3 and R version 3.6.2.

Results

Patient Characteristics

A total of 357 samples from 282 patients were included in the current study: 234 patients were tested once, 31 were tested twice, 11 were tested three times, 5 were tested four times, and 1 was tested eight times. Demographic data were available for 282 patients and are summarized in supplemental online Table 1. A subset of clinical data and outcomes for samples collected prior to first‐line treatment in 104 patients with locally advanced or metastatic PDAC is presented in Table 1.

Landscape of Alterations

Ninety percent of the 357 ctDNA samples harbored at least 1 genomic alteration, including VUS. Each sample harbored an average of 2.7 alterations, including SNVs, indels, and/or amplifications, with a median allelic fraction of 0.40%. The overall landscape of unique alterations in each patient is summarized in Figure 1, depicted by PDAC. TP53 and KRAS were the two most common alterations, followed by SMAD4, CDKN2A, and EGFR. This histogram also displays the frequencies of alterations detected in tissue samples from patients with PDAC using the publicly available cBioPortal database for cancer genomics.

Figure 1.

Figure 1

Open in a new tab

Frequency of alterations identified in patients with PDAC with tissue‐based next‐generation sequencing versus circulation tumor DNA samples. Abbreviation: PDAC, pancreatic ductal adenocarcinoma.

An oncoprint of the genomic landcape among all PDAC samples is presented in Figure 2. A total of 52 Chromosomal copy number variations were identified in 25 samples and most frequently involved MYC (32%), CDK6 (28%), KRAS (28%), FGFR1 (20%), EGFR (16%), PDGFRA (12%), and PIK3CA (12%). The distribution of plasma copy number is presented in Figure 3.

Figure 2.

Figure 2

Open in a new tab

Oncoprint demonstrating the landscape of co‐occuring alterations in the pancreatic ductal adenocarcinoma cohort.

Figure 3.

Figure 3

Open in a new tab

Distribution of plasma copy number in the most frequently observed gene amplifications. The dotted line represents a normal copy number of 2.0.

Therapeutically Relevant Alterations

After excluding VUS, therapeutically relevant alterations were observed in 170 (48%) of the total cohort of samples. Supplemental online Figure 1 summarizes the 10 most commonly altered therapeutically relevant genes. KRAS (88%) was the most frequently identified therapeutically relevant alteration, followed by PIK3CA (7%), ATM (5.3%), EGFR (5.3%), and MYC (5.3%) mutations. Homologous recombination gene mutations (HRm) were detected in approximately 8.8% of the patients (BRCA1/2 3.5%; ATM 5.3%). Some of these HRm were detected at a variant allele frequency suspicious for a germline origin and subsequently confirmed on germline testing, which was confirmed in two patients [22].

The most common co‐occuring mutations were found in TP53 and KRAS (n = 74; p < .001), KRAS and SMAD4 (n = 25; p < .001), KRAS and CDKN2A (n = 12; p = .134), and TP53 and SMAD4 (n = 20; p = .024). The distribution of KRAS point mutations has been illustrated in Figure 4, with the most common variants being G12D (n = 55), G12V (n = 51), G12R (n = 28), and G12C (n = 4), making up 36%, 33%, 18%, and 2.6% of all KRAS mutations, respectively.

Figure 4.

Figure 4

Open in a new tab

Distribution of KRAS variants identified in the cohort. Abbreviation: aa, amino acid.

The graph in Figure 5 shows the difference in frequency of the detected alterations from samples collected at baseline (treatment naive), while on treatment, and at disease progression. KRAS, SMAD, and BRCA1 mutations were identified more frequently in patients at baseline and disease progression, whereas CCNE mutations were more frequently observed at disease progression, and TP53 and BRCA2 mutations occurred most frequently in patients who were on treatment.

Figure 5.

Figure 5

Open in a new tab

Frequency of genetic alterations at different stages of treatment (initial, middle, and progression).

A subset of 104 patients with baseline ctDNA results available prior to initiation of system therapy were analyzed to evaluate the presence or absence of impactful genes with either locally advanced or metastatic disease. The presence of KRAS, CCND2, SMAD, and/or TP53 alterations at baseline was significantly associated with metastatic disease (Table 2). Metastatic disease had a median of three somatic alterations per sample compared with a median of one somatic alteration per sample in locally advanced disease (p < .001). In patients with known response to chemotherapy, only 10 of 48 (28%) patients with KRAS had a response as compared with 26 of 42 (72%) of patients without a KRAS mutation (p < .001). No other mutations were associated with response. By univariate analysis, the presence of three or more genetic alterations, presence of KRAS mutations, and presence of TP53 mutations were significantly associated with OS. However, these genes did not show a significant association with OS in the multivariate analysis (supplemental online Table 2).

Table 2.

Association of genetic alterations to the stage of disease at baseline

Genetic alterations Locally advanced (n = 39), n (%) Metastatic (n = 65), n (%) Total (n = 104), n (%) p value
HRD (BRCA1/2 or ATM) .459 a
Absent 36 (92.3) 57 (87.7) 93 (89.4)
Present 3 (7.7) 8 (12.3) 11 (10.6)
CCND2 .037 a
Absent 37 (94.9) 52 (80.0) 89 (85.6)
Present 2 (5.1) 13 (20.0) 15 (14.4)
SMAD .015 a
Absent 39 (100.0) 56 (86.2) 95 (91.3)
Present 0 (0.0) 9 (13.8) 9 (8.7)
KRAS <.001 a
Absent 28 (71.8) 19 (29.2) 47 (45.2)
Present 11 (28.2) 46 (70.8) 57 (54.8)
KRAS category <.001 a
Absent 28 (71.8) 19 (29.2) 47 (45.2)
Yes: 1 b 11 (28.2) 41 (63.1) 52 (50.0)
Yes: 2+ b 0 (0.0) 5 (7.7) 5 (4.8)
TP53 .010 a
Absent 22 (56.4) 20 (30.8) 42 (40.4)
Present 17 (43.6) 45 (69.2) 62 (59.6)
TP53 category .013 a
Absent 22 (56.4) 20 (30.8) 42 (40.4)
Yes: 1 b 15 (38.5) 31 (47.7) 46 (44.2)
Yes: 2+ b 2 (5.1) 14 (21.5) 16 (15.4)
PIK3CA .447 a
Absent 35 (89.7) 61 (93.8) 96 (92.3)
Present 4 (10.3) 4 (6.2) 8 (7.7)
ATM .613 a
Absent 37 (94.9) 90 (92.3) 97 (93.3)
Present 2 (5.1) 5 (7.7) 7 (6.7)
Presence of genes alterations <.001 a
No alterations 16 (41.0) 8 (12.3) 24 (23.1)
Yes alterations 23 (59.0) 57 (87.7) 80 (76.9)
Number of genetic alterations <.001 c
Count 39 65 104
Median (range) 1.0 (0.0–6.0) 3.0 (0.0–12.0) 2.0 (0.0–12.0)
3+ genetic alterations <.001 a
Absent 33 (84.6) 29 (44.6) 62 (59.6)
Present 6 (15.4) 36 (55.4) 42 (40.4)
Open in a new tab
a

Pearson's chi‐square test.

b

Number of the genetic location.

c

Kruskal‐Wallis rank sum test.

Among 282 patients, 40 patients had at least two serial ctDNA assays performed at the time of diagnosis and disease progression. A total of 23 of 40 (57.5 %) patients acquired new genetic alterations that were identified by ctDNA at disease progression: 7 of 40 (17.5%) acquired VUS alterations, and 16 of 40 (40%) acquired pathogenic alterations. None of these newly acquired genetic alterations were identified on tissue profiling performed at diagnosis.

Therapeutically relevant alterations were identified in 12 of the 40 (30 %) patients, with EGFR (7.5%), PIK3CA (5.0%), RET (5.0%), MET (5.0%), BRCA1 (2.5%), PDGFRA (2.5%), ERBB2 (2.5%), and FGFR2 (2.5%) most commonly acquired at progression. These patients received second‐line regimen in the form of chemotherapy upon progression, and none of these patients were enrolled in a clinical trial targeting these newly aquired alterations/pathways.

Tissue NGS was performed on 57 of 171 patients with available tissue biopsy (33%) at diagnosis, with a median of 130 days between tissue biopsy and ctDNA analysis. High concordance was identified between tissue and ctDNA NGS for various biomarkers including TP53 (86%), KRAS (81%), and other therapeutically relevant genes such as FGFR, BRCA1, BRCA2, ERBB2, STK11, and CDKN2A (50%; supplemental online Figure 2).

Discussion

In this study, we show the feasibility of ctDNA testing using a 73‐gene panel and identifying therapeutically relevant targets in 48% of patients with PDAC. Most frequently observed targets included KRAS, ATM, MYC, PIK3CA, and EGFR. In addition, HRm were detected in 8.5% of the patients. KRAS, SMAD, and BRCA1 mutations were identified more frequently in patients at baseline and at disease progression. Cyclin E1(CCNE) mutations were more frequently observed at disease progression, suggesting that these genes are likely to be acquired as potential resistance alterations to therapy. TP53 and BRCA2 mutations occurred most frequently in patients who were on treatment, suggesting that the frequency of alterations differs by the timing of ctDNA blood draw.

Historically, molecular profiling was solely done on surgically resected or biopsied tissue. Although surgically obtained tissue is ideal, the majority of patients with PDAC are not surgical candidates. In patients who are not surgical candidates, biopsies of the primary tumor or metastatic disease are obtained. However, the availability of sufficient tumor cell quantity is often inadequate from biopsied samples. Furthermore, PDACs demonstrate intratumor molecular heterogeneity, and obtaining small tissue samples may not provide an adequate representation of the entire molecular status [23, 24, 25].

With the availability of next‐generation genomic profiling in recent years, the genomic landscape of PDAC has emerged, with real effects on our understanding of the biology of this disease and with the realization of a number of therapeutically relevant targets. Next‐generation genomic profiling using ctDNA has emerged as a potentially powerful method to evaluate a patient's tumor's genomic landscape. Investigations of the use of ctDNA in PDAC have previously been limited by the relatively incomplete understanding of the genetics of this cancer and the lack of a validated and widely used ctDNA assay. This has now changed with recent increased focus on the disease; recent characterization of the PDAC genome by several studies, including that from The Cancer Genome Atlas; and the availability of a number of commercially available ctDNA assays in clinical practice [26]. Studies are limited in this area, but thus far, data have revealed tissue‐derived DNA and ctDNA are comparable with high concordance rates, although ctDNA has shown a better representation of the heterogeneity in PDAC [3, 23, 27].

Vietsch et al. (2017) [23] compared ctDNA with surgical specimens of primary tumors in five patients with PDAC using a 56‐gene panel. They showed 78% of the mutations identified via ctDNA were detected in tumor tissue DNA, demonstrating that ctDNA may provide a more dynamic profile with more potentially therapeutic targets [23]. This suggests that because of intratumor heterogeneity and primary tumor section sampling, tissue molecular profiling may fail to adequately profile the tumor whereas ctDNA may capture more oncogenic mutations. Studies using ctDNA in PDAC have recently been gaining popularity; however, those published have been limited by sample size and/or limited gene sets or have focused on clinical associations of specific targets with outcome [27, 28, 29, 30, 31].

Many of these targets have prognostic and potentially predictive therapeutic implications for prognosis in PDAC. Recent studies using deep sequencing of ctDNA were able to identify therapeutic targets in 29% of patients [32]. KRAS mutation is observed in >90% of PDAC and recently, there have been some breakthroughs that have exploited a specific subset of KRAS as a possible therapeutic target. One clinical study with AMG 510, an agent targeting KRAS G12C, included four patients with PDAC, of whom two responded with a > 50% tumor shrinkage, including one with a complete response [33]. Another selective covalent KRAS G12C compound, MRTX849, appears to be equally interesting and is currently being developed [34]. Additionally, the presence of KRAS mutations have been shown to correlate significantly with PFS and OS in both locally advanced and metastatic PDAC [27, 28, 29, 30, 31, 35, 36, 37, 38, 39].

Evidence indicates that patients with PDAC harboring HRm have better response to platinum‐based chemotherapy [6, 40] and may also respond to poly adenosine phosphate‐ribose polymerase (PARP) inhibitors [4, 41]. Most recently, the randomized phase III trial (POLO) showed better PFS with olaparib (a PARP inhibitor) in patients with PDAC with germline BRCA mutation [4]. We were not able to determine PARP inhibitor efficacy in our study because of a small sample size and most of the patients received gemcitabine based regimen. In addition, a recent randomized study showed that that patients with germline BRCA1/2‐ or PALB2‐mutated PDAC yielded very high response rates on first‐line chemotherapy regimen [40]. This was recently shown in a retrospective study of 262 patients with PDAC in which 19% of the patients were found to have HRm (defined as germline or somatic pathogenic alterations of 17 HRm including ATM, BRCA1, and BRCA2); PFS was better in patients with HRm when platinum‐based therapy was used compared with nonplatinum‐based therapy [6]. Additional series are corroborating better outcomes in patients harboring the HRm aberrations, highlighting the value of identifiyin these patients [42]. In our study, we detected HRm in 8.8% of the patients. These findings could potentially help physicians in tailoring first‐line regimens by choosing FOLFIRINOX or combining gemcitabine with cisplatin instead of the more commonly used nab‐paclitaxel. This is especially important as the turn‐around time for ctDNA is usually much faster than a tissue‐based testing.

Conclusion

Although our study includes the largest sample size using the largest gene panel thus far in PDAC, it is limited by its retrospective nature and the limited amount of clinical data available for 165 of the 282‐patient cohort. Given the recent data showing mutation alterations with PDAC progression [31], it would be important to analyze these patients for changes in ctDNA targets during the course of their disease and treatments. Future studies should investigate the frequency necessary for serial ctDNA testing in the course of disease, optimal ctDNA quantity, the utility of ctDNA in tailoring individualized therapy in PDAC.

Finally, ct‐DNA testing platforms are now also reporting out plasma‐microsatelite instability as part of their reports [43]. This has been validated and although present in <1% of PDAC, for the ones who have it, immunotherapy is a very viable and durable, and potentially curable, option [44].

Author Contributions

Conception/design: Gehan Botrus, Kabir Mody, Tanios S. Bekaii‐Saab

Provision of study material or patients: Gehan Botrus, Mohamad Bassam Sonbol, Mitesh J. Borad, Daniel H. Ahn, Phani Keerthi Surapaneni, Jason Starr, Ashton Boyle, Jessica McMillan, Natasha Wylie, Kabir Mody, Tanios S. Bekaii‐Saab

Collection and/or assembly of data: Gehan Botrus, Bassam Sonbol

Data analysis and interpretation: Gehan Botrus, Bassam Sonbol, Leylah M. Drusbosky

Manuscript writing: Gehan Botrus, Bassam Sonbol, Yael Kusne, Kabir Mody

Final approval of manuscript: Gehan Botrus, Heidi Kosirorek, Mohamad Bassam Sonbol, Yael Kusne, Pedro Luiz Serrano Uson Junior, Mitesh J. Borad, Daniel H. Ahn, Kasi M. Pashtoon, Leylah M. Drusbosky, Hiba Dada, Phani Keerthi Surapaneni, Jason Starr, Ashton Ritter, Jessica McMillan, Natasha Wylie, Kabir Mody, Tanios S. Bekaii‐Saab

Disclosures

Mitesh J. Borad: Senhwa Pharmaceuticals, Adaptimmune, Agios Pharmaceuticals, Halozyme Pharmaceuticals, Celgene Pharmaceuticals, EMD Merck Serono, Toray, Dicerna, Taiho Pharmaceuticals, Sun Biopharma, Isis Pharmaceuticals, Redhill Pharmaceuticals, Boston Biomed, Basilea, Incyte Pharmaceuticals, Mirna Pharmaceuticals, Medimmune, Bioline, Sillajen, ARIAD Pharmaceuticals, PUMA Pharmaceuticals, Novartis Pharmaceuticals, QED Pharmaceuticals, Pieris Pharmaceuticals (RF), ADC Therapeutics, Exelixis Pharmaceuticals, Inspyr Therapeutics, G1 Therapeutics, Immunovative Therapies, OncBioMune Pharmaceuticals, Western Oncolytics, Lynx Group, Genentech, Merck, Huya (C/A); Daniel H. Ahn: Exelixis, Genentech, Eisai, AstraZeneca, Bayer (C/A); Kasi M. Pashtoon: Natera, Foundation Medicine, Merck, Taiho, Ipsen (C/A), Boston Scientific, AstraZeneca, Tersera (RF); Jason Starr: Ipsen, Natera, Pfizer (C/A), Rafael, Incyte, Merus, Aprea, Cardiff, Molecular Templates, Macrogenics, Daichii (RF); Tanios S. Bekaii‐Saab: Abgenomics, Amgen, Array Biopharma, Bayer, Bristol‐Myers Squibb, Boston Biomedical, Celgene, Clovis, Genentech, Incyte, Ipsen, Lilly, Merck, Seattle Genetics (RF), Array Biopharma, Bayer, Boehringer Ingelheim, Genentech, Incyte, Ipsen, Merck, TreosBio, Seattle Genetics, Sobi (H), AstraZeneca, Eli Lilly & Co, PanCan, 1Globe, Sun Biopharma, Imugene, Immuneering (Other), patents WO/2018/183488 and WO/201/055687 (IP). The other authors indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board

Supporting information

See http://www.TheOncologist.com for supplemental material available online.

Figure S1 A Top 10 Therapeutically Relevant Alterations identified in cDNA of PDAC patients.

Figure S2: Clonal Concordance rates

Click here for additional data file. (121.8KB, pdf)

Table S1 Patient demographics

Table S2 OS univariate and multivariable analysis

Click here for additional data file. (143.4KB, pdf)

Acknowledgments

No funding was used to support this study. This work was presented in part as a poster at the American Society of Clinical Oncology 2019 Gastrointestinal Cancers Symposium.

No part of this article may be reproduced, stored, or transmitted in any form or for any means without the prior permission in writing from the copyright holder. For information on purchasing reprints contact commercialreprints@wiley.com. For permission information contact permissions@wiley.com.

Disclosures of potential conflicts of interest may be found at the end of this article.

References

  • 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin 2018;68:7–30. [DOI] [PubMed] [Google Scholar]
  • 2. Poruk KE, Firpo MA, Adler DG et al. Screening for pancreatic cancer: Why, how, and who? Ann Surg 2013;257:17–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Wei T, Zhang Q, Li X et al. Monitoring tumor burden in response to FOLFIRINOX chemotherapy via profiling circulating cell‐free DNA in pancreatic cancer. Mol Cancer Ther 2019;18:196–203. [DOI] [PubMed] [Google Scholar]
  • 4. Golan T, Hammel P, Reni M et al. Maintenance olaparib for germline BRCA‐mutated metastatic pancreatic cancer. N Engl J Med 2019;381:317–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Wattenberg MM, Asch D, Yu S et al. Platinum response characteristics of patients with pancreatic ductal adenocarcinoma and a germline BRCA1, BRCA2 or PALB2 mutation. Br J Cancer 2020;122:333–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Park W, Chen J, Chou JF et al. Genomic methods identify homologous recombination deficiency in pancreas adenocarcinoma and optimize treatment selection. Clin Cancer Res 2020;26:3239–3247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. McAnena P, Brown JA, Kerin MJ. Circulating nucleosomes and nucleosome modifications as biomarkers in cancer. Cancers (Basel) 2017;9:5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Lissa D, Robles AI. Methylation analyses in liquid biopsy. Transl Lung Cancer Res 2016;5:492–504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Lu L, Bi J, Bao L. Genetic profiling of cancer with circulating tumor DNA analysis. J Genet Genomics 2018;45:79–85. [DOI] [PubMed] [Google Scholar]
  • 10. Warton K, Samimi G. Methylation of cell‐free circulating DNA in the diagnosis of cancer. Front Mol Biosci 2015;2:13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Newman AM, Bratman SV, To J. An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage. Nat Med 2014;20:548–554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Wang Z, Cheng Y, An T et al. Detection of EGFR mutations in plasma circulating tumour DNA as a selection criterion for first‐line gefitinib treatment in patients with advanced lung adenocarcinoma (BENEFIT): A phase 2, single‐arm, multicentre clinical trial. Lancet Respir Med 2018;6:681–690. [DOI] [PubMed] [Google Scholar]
  • 13. Rothwell DG, Ayub M, Cook N et al. Utility of ctDNA to support patient selection for early phase clinical trials: the TARGET study. Nat Med 2019;25:738–743. [DOI] [PubMed] [Google Scholar]
  • 14. Tuaeva NO, Falzone L, Porozov YB et al. Translational application of circulating DNA in oncology: Review of the last decades achievements. Cells 2019;8:1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Howes N, Lerch MM, Greenhalf W et al. Clinical and genetic characteristics of hereditary pancreatitis in Europe. Clin Gastroenterol Hepatol 2004;2:252–261. [DOI] [PubMed] [Google Scholar]
  • 16. Von Hoff DD, Ramanathan RK, Borad MJ et al. Gemcitabine plus nab‐paclitaxel is an active regimen in patients with advanced pancreatic cancer: A phase I/II trial. J Clin Oncol 2011;29:4548–4554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Conroy T, Desseigne F, Ychou M et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 2011;364:1817–1825. [DOI] [PubMed] [Google Scholar]
  • 18. Lanman RB, Mortimer SA, Zill OA et al. Analytical and clinical validation of a digital sequencing panel for quantitative, highly accurate evaluation of cell‐free circulating tumor DNA. PLoS One 2015;10:e0140712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Odegaard JI, Vincent JJ, Mortimer S et al. Validation of a plasma‐based comprehensive cancer genotyping assay utilizing orthogonal tissue‐ and plasma‐based methodologies. Clin Cancer Res 2018;24:3539–3549. [DOI] [PubMed] [Google Scholar]
  • 20. Zill OA, Banks KC, Fairclough SR et al. The Landscape of Actionable Genomic Alterations in Cell‐Free Circulating Tumor DNA from 21,807 Advanced Cancer Patients. Clin Cancer Res 2018;24:3528–3538. [DOI] [PubMed] [Google Scholar]
  • 21. Chakravarty D, Gao J, Phillips SM et al. OncoKB: A precision oncology knowledge base. JCO Precis Oncol 2017;2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Slavin TP, Banks KC Chudova D et al. Identification of incidental germline mutations in patients with advanced solid tumors who underwent cell‐free circulating tumor DNA sequencing. J Clin Oncol 2018;36:JCO1800328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Vietsch EE, Graham GT, McCutcheon JN et al. Circulating cell‐free DNA mutation patterns in early and late stage colon and pancreatic cancer. Cancer Genet 2017;218–219:39–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Nakamura T, Kuwai T, Kitadai Y et al. Zonal heterogeneity for gene expression in human pancreatic carcinoma. Cancer Res 2007;67:7597–7604. [DOI] [PubMed] [Google Scholar]
  • 25. Harada T, Okita K, Shiraishi K et al. Interglandular cytogenetic heterogeneity detected by comparative genomic hybridization in pancreatic cancer. Cancer Res 2002;62:835–839. [PubMed] [Google Scholar]
  • 26. Cancer Genome Atlas Research Network; Cancer Genome Atlas Research . Integrated genomic characterization of pancreatic ductal adenocarcinoma. Cancer Cell 2017;32:185–203.e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Bernard V, Kim DU, San Lucas FA et al. Circulating nucleic acids are associated with outcomes of patients with pancreatic cancer. Gastroenterology 2019;156:108–118.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Allenson K, Castillo J, San Lucas FA et al. High prevalence of mutant KRAS in circulating exosome‐derived DNA from early‐stage pancreatic cancer patients. Ann Oncol 2017;28:741–747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Adamo P, Cowley CM, Neal CP et al. Profiling tumour heterogeneity through circulating tumour DNA in patients with pancreatic cancer. Oncotarget 2017;8:87221–87233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Mohan S, Ayub M, Rothwell DG et al. Analysis of circulating cell‐free DNA identifies KRAS copy number gain and mutation as a novel prognostic marker in pancreatic cancer. Sci Rep 2019;9:11610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Perets R, Greenberg O, Shentzer T et al. Mutant KRAS Circulating tumor DNA is an accurate tool for pancreatic cancer monitoring. The Oncologist 2018;23:566–572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Takai E, Totoki Y, Nakaruma H et al. Clinical utility of circulating tumor DNA for molecular assessment in pancreatic cancer. Sci Rep 2015;5:18425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Canon J, Rex K, Saiki AY et al. The clinical KRAS(G12C) inhibitor AMG 510 drives anti‐tumour immunity. Nature 2019;575:217–223. [DOI] [PubMed] [Google Scholar]
  • 34. Hallin J, Engstrom LD, Hargis L et al. The KRAS(G12C) inhibitor MRTX849 provides insight toward therapeutic susceptibility of KRAS‐mutant cancers in mouse models and patients. Cancer Discov 2020;10:54–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Chen I, Raymond VM, Geis JA et al. Ultrasensitive plasma ctDNA KRAS assay for detection, prognosis, and assessment of therapeutic response in patients with unresectable pancreatic ductal adenocarcinoma. Oncotarget 2017;8:97769–97786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Del Re M, Vivaldi C, Rofi E et al. Early changes in plasma DNA levels of mutant KRAS as a sensitive marker of response to chemotherapy in pancreatic cancer. Sci Rep 2017;7:7931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Kruger S, Heinemann V, Ross C et al. Repeated mutKRAS ctDNA measurements represent a novel and promising tool for early response prediction and therapy monitoring in advanced pancreatic cancer. Ann Oncol 2018;29:2348–2355. [DOI] [PubMed] [Google Scholar]
  • 38. Tjensvoll K, Lapin M, Buhl T et al. Clinical relevance of circulating KRAS mutated DNA in plasma from patients with advanced pancreatic cancer. Mol Oncol 2016;10:635–643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Sugimori M, Sugimori K, Tsuchiya H et al. Quantitative monitoring of circulating tumor DNA in patients with advanced pancreatic cancer undergoing chemotherapy. Cancer Sci 2020;111:266–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. O'Reilly EM, Lee JW, Zalupski M et al. Randomized, multicenter, phase II trial of gemcitabine and cisplatin with or without veliparib in patients with pancreas adenocarcinoma and a germline BRCA/PALB2 mutation. J Clin Oncol 2020;38:1378–1388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Pishvaian MJ, Blais EM, Brody JR et al. Overall survival in patients with pancreatic cancer receiving matched therapies following molecular profiling: A retrospective analysis of the Know Your Tumor registry trial. Lancet Oncol 2020;21:508–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Macklin‐Mantia SK, Hines SL, Kasi PM. Retrospective review of outcomes in patients with DNA‐damage repair related pancreatic cancer. Hered Cancer Clin Pract 2020;18:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Willis J, Lefterova MI, Artyomenko A et al. Validation of microsatellite instability detection using a comprehensive plasma‐based genotyping panel. Clin Cancer Res 2019;25:7035–7045. [DOI] [PubMed] [Google Scholar]
  • 44. Kamatham S, Shahjehan F, Kasi PM. Circulating tumor DNA‐based detection of microsatellite instability and response to immunotherapy in pancreatic cancer. Front Pharmacol 2020;11:23. [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

See http://www.TheOncologist.com for supplemental material available online.

Figure S1 A Top 10 Therapeutically Relevant Alterations identified in cDNA of PDAC patients.

Figure S2: Clonal Concordance rates

Click here for additional data file. (121.8KB, pdf)

Table S1 Patient demographics

Table S2 OS univariate and multivariable analysis

Click here for additional data file. (143.4KB, pdf)

Articles from The Oncologist are provided here courtesy of Oxford University Press

RESOURCES