Pancreatic Cancer
ctDNA Monitoring and Molecular Profiling in Oncology's Most Challenging Disease
Clinical Overview
Pancreatic ductal adenocarcinoma (PDAC) remains one of oncology's greatest challenges, with a 5-year survival rate of approximately 12%. Even after curative-intent surgery, over 80% of patients develop recurrence, typically within the first 2 years. Early detection of recurrence and identification of actionable molecular targets are critical, as treatment options become increasingly limited with disease progression.
ctDNA testing offers two complementary applications in pancreatic cancer: minimal residual disease (MRD) monitoring to detect early recurrence post-resection, and molecular profiling to identify targetable mutations that may guide therapy selection. However, pancreatic cancer presents unique biological challenges that affect ctDNA performance.
Why ctDNA Matters in Pancreatic Cancer
- Early recurrence detection: 4.5-6.5 months before radiographic progression
- Prognostic stratification: ctDNA-positive patients have 3.9-12.2 times higher recurrence risk
- Actionable genotyping: Identifies KRAS variants (including emerging G12D inhibitors), BRCA1/2 mutations eligible for PARP inhibitors, and rare but actionable targets
- Non-invasive profiling: Avoids risks of pancreatic biopsy in critically ill or surgically unresectable patients
- Superior to CA19-9: ctDNA shows stronger prognostic value than traditional tumor marker
ctDNA Testing Methodology
ctDNA testing in pancreatic cancer employs targeted sequencing approaches to detect tumor-derived DNA fragments in blood plasma. Two primary strategies are used:
Tumor-Informed (Baseline-Based) Approach:
- Uses a baseline sample (tissue biopsy or baseline plasma) to identify the patient's specific tumor mutations
- Subsequent monitoring tracks these identified mutations at serial timepoints
- Higher analytical sensitivity for MRD detection in the post-resection setting
- Requires baseline tissue or pre-treatment blood sample
Tumor-Agnostic (No Baseline) Approach:
- Tests directly at monitoring timepoints without prior baseline profiling
- Uses fixed gene panels targeting common pancreatic cancer mutations (KRAS, TP53, SMAD4, CDKN2A)
- Suitable for molecular profiling to identify actionable targets
- No requirement for baseline sample, can be performed at any disease stage
Clinical Consideration: Tumor-informed approaches generally achieve higher sensitivity for MRD detection, while tumor-agnostic approaches are practical for genotyping in metastatic disease or when baseline samples are unavailable. Both approaches can use fixed gene panels; the key distinction is whether baseline profiling was performed.
MRD Detection: Clinical Utility and Limitations
Clinical Context: After surgical resection of pancreatic cancer, detecting minimal residual disease can identify patients at highest risk for recurrence. However, pancreatic cancer is a "low-shedder" tumor type with unique biological challenges that impact ctDNA detection performance.
MRD Detection Performance Metrics:
- Sensitivity: 43-71% post-resection (significantly lower than colorectal cancer at 80-90%)
- Lead Time: 4.5-6.5 months before radiographic recurrence detection
- Hazard Ratio for Recurrence: 3.9-12.2 for ctDNA-positive vs ctDNA-negative patients
- Comparison to CA19-9: ctDNA demonstrates superior prognostic value (HR 7.8 vs HR 1.9 for CA19-9)
Interpreting the Hazard Ratios: The wide range of hazard ratios (3.9-12.2) across studies reflects differences in patient populations, timing of blood collection, and assay methodologies. Even at the lower end (HR 3.9), ctDNA-positive patients face approximately 4 times the risk of recurrence compared to ctDNA-negative patients. Higher hazard ratios (HR 7.8-12.2) in some studies suggest that ctDNA may identify a subset of patients with particularly aggressive biology.
The "Low-Shedder" Challenge: Biological Limitations of ctDNA in Pancreatic Cancer
Pancreatic cancer presents unique biological challenges that limit ctDNA detection sensitivity:
- Desmoplastic Stroma: Dense fibrous tissue surrounding pancreatic tumors impairs blood vessel access, reducing ctDNA shedding into circulation
- Rapid Hepatic Clearance: The pancreas drains directly into the portal venous system, allowing the liver to rapidly clear ctDNA before it reaches systemic circulation
- Lower Tumor Burden Post-Resection: Microscopic residual disease sheds far less ctDNA than bulky tumors, making detection more challenging
- Stage-Dependent Detection: Early-stage disease has lower detection rates than advanced disease, limiting utility in the curative setting where MRD monitoring is most needed
Clinical Implication: These biological factors result in sensitivity of 43-71%, meaning that approximately 30-57% of patients with actual residual disease will have undetectable ctDNA. This false-negative rate is substantially higher than in other gastrointestinal cancers.
Key Clinical Studies
Groot et al. (Clin Cancer Res 2019): Prospective tumor-informed study of 112 resected PDAC patients:
- Detection of ctDNA post-surgery strongly predicted recurrence (HR 12.2, p<0.001)
- Median lead time of 6.5 months before radiographic recurrence
- Longitudinal monitoring (serial timepoints) improved detection rates over single assessment
- ctDNA superior to CA19-9 for recurrence prediction (HR 7.8 vs 1.9)
Bernard et al. (Gastroenterology 2019): Multi-marker approach in 194 patients:
- Postoperative ctDNA associated with worse overall survival (HR 3.7, 95% CI 2.0-6.7, p<0.001)
- ctDNA dynamics during adjuvant therapy predicted clinical outcomes
- Combined analysis of ctDNA and CA19-9 improved prognostic accuracy beyond either marker alone
Watanabe et al. (Ann Surg Oncol 2019): Longitudinal monitoring study:
- ctDNA detection predicted recurrence with median lead time of 4.5 months
- Sensitivity improved with serial measurements vs single timepoint (71% vs 52%)
- Negative ctDNA did not exclude recurrence (false-negative rate ~29%)
References: Groot et al. Clin Cancer Res 2019, Bernard et al. Gastroenterology 2019, Watanabe et al. Ann Surg Oncol 2019
Molecular Profiling: Actionable Targets and Therapeutic Options
1. KRAS Mutations: Expanding Therapeutic Landscape
Clinical Context: KRAS is the most commonly mutated oncogene in pancreatic cancer, present in over 90% of cases. For decades, KRAS was considered "undruggable," but recent breakthroughs have produced mutation-specific inhibitors. Identifying the specific KRAS variant is now clinically actionable.
KRAS Variant Distribution in Pancreatic Cancer:
- KRAS G12D: ~40% of all cases (most common variant)
- KRAS G12V: ~30% (second most common)
- KRAS G12R: ~15-20%
- KRAS G12C: 1-2% (currently targetable with approved drugs)
- Other variants: G12A, G12S, Q61H (remainder)
KRAS G12C Inhibitors: Available Targeted Therapy
Sotorasib (CodeBreaK 100 Trial):
- Patient Population: KRAS G12C-mutant advanced solid tumors, including pancreatic cancer
- Overall Response Rate (ORR): 26.7% in KRAS G12C-mutant PDAC
- Disease Control Rate: 84% (includes partial responses + stable disease)
- Clinical Limitation: Only 1-2% of pancreatic cancer patients harbor G12C mutation
Adagrasib:
- Second-generation KRAS G12C inhibitor with clinical activity in PDAC
- May have blood-brain barrier penetration (relevant for brain metastases)
- Clinical efficacy data similar to sotorasib in G12C-mutant solid tumors
Clinical Application: While only 1-2% of pancreatic cancer patients have G12C mutations, identifying these patients through ctDNA profiling provides access to approved targeted therapy with meaningful response rates. ctDNA testing is particularly valuable when tissue biopsy is not feasible.
KRAS G12D Inhibitors: Emerging Therapy with Major Impact Potential
Clinical Significance: G12D is the most common KRAS variant in pancreatic cancer (~40% of cases), making G12D-specific inhibitors a clinically significant therapeutic advance.
MRTX1133 and Other G12D Inhibitors:
- Target Population: ~40% of pancreatic cancer patients with KRAS G12D mutations
- Development Stage: Multiple G12D-specific inhibitors in Phase I/II clinical trials
- Mechanism: Selectively target the G12D mutant protein, sparing wild-type KRAS
- Early Data: Preclinical models demonstrate potent anti-tumor activity; clinical trial results emerging
Clinical Implication: If KRAS G12D inhibitors achieve regulatory approval, they would represent a major therapeutic advance for the single largest molecular subgroup of pancreatic cancer patients. ctDNA genotyping would become essential for identifying eligible patients, particularly in settings where tissue biopsy is not feasible or when tissue is exhausted from prior molecular testing.
Current Recommendation: Patients with advanced pancreatic cancer should undergo KRAS variant testing (tissue or ctDNA) to identify G12D mutations for clinical trial enrollment and to prepare for potential future approved therapies.
References: Strickler et al. N Engl J Med 2023, Bekaii-Saab et al. J Clin Oncol 2023
2. BRCA1/2 Mutations: PARP Inhibitor Therapy
Clinical Context: BRCA1/2 mutations impair homologous recombination DNA repair, creating synthetic lethality with PARP inhibitors. Both germline (inherited) and somatic (tumor-acquired) mutations are therapeutically relevant.
BRCA Mutations in Pancreatic Cancer:
- Germline BRCA1/2: 4-7% of all PDAC patients
- Somatic BRCA1/2: Additional ~3-5% acquired in tumor tissue
- Total Actionable Population: ~7-12% eligible for PARP inhibitor therapy
- Other HR Deficiency Genes: PALB2, ATM, RAD51C/D may also predict PARP inhibitor sensitivity
POLO Trial: Olaparib Maintenance Therapy (Level 1 Evidence)
The landmark POLO trial established olaparib as standard-of-care maintenance therapy for germline BRCA-mutated metastatic pancreatic cancer following platinum-based chemotherapy:
Study Design:
- Phase 3 randomized controlled trial
- Patient population: Germline BRCA1/2-mutated metastatic PDAC without disease progression after ≥16 weeks of platinum-based chemotherapy
- Treatment: Olaparib 300 mg twice daily vs placebo (maintenance setting)
Efficacy Results:
- Median Progression-Free Survival: 7.4 months (olaparib) vs 3.8 months (placebo) [HR 0.53, 95% CI 0.35-0.82, p=0.004]
- Disease Control Rate: 85% with olaparib vs 62% with placebo
- Overall Survival: Trend toward improvement (18.9 vs 18.1 months, not statistically significant)
- Clinical Benefit: Near-doubling of progression-free survival with maintenance olaparib
Clinical Application: ctDNA testing can identify both germline and somatic BRCA mutations, potentially expanding the eligible population beyond germline testing alone. Somatic BRCA mutations detected in ctDNA may also predict PARP inhibitor sensitivity, though prospective validation is ongoing. ctDNA profiling is particularly valuable when tissue is unavailable or insufficient for comprehensive molecular analysis.
Reference: Golan et al. N Engl J Med 2019
3. Rare but Actionable Targets
Clinical Context: While less common, several additional molecular alterations in pancreatic cancer have matched targeted therapies.
Microsatellite Instability-High (MSI-H) / Mismatch Repair Deficient (dMMR):
- Prevalence: 1-2% of pancreatic cancers
- Therapy: Pembrolizumab (anti-PD-1 immunotherapy)
- Response Rate: ~30% ORR across MSI-H solid tumors
- Clinical Significance: Immunotherapy highly effective in this rare subgroup, with durable responses
NTRK Fusions:
- Prevalence: <1% of pancreatic cancers
- Therapy: Larotrectinib, entrectinib (TRK inhibitors)
- Response Rate: ~75% ORR across NTRK fusion-positive solid tumors
- Clinical Significance: Dramatic responses in rare fusion-positive cases
HER2 Amplification:
- Prevalence: ~2% of pancreatic cancers
- Therapy: Trastuzumab deruxtecan (antibody-drug conjugate) - clinical trials ongoing
- Early Data: Promising activity in HER2-amplified pancreatic cancer in early-phase trials
- Clinical Significance: Emerging therapeutic option for small molecular subgroup
Clinical Implication: Comprehensive molecular profiling through ctDNA or tissue sequencing is essential to identify these rare but highly actionable alterations. Even at low prevalence rates, identification of MSI-H, NTRK fusions, or HER2 amplification can fundamentally change treatment strategy and prognosis.
Ongoing Clinical Trials
Clinical Context: Multiple interventional trials are testing whether acting on ctDNA results improves patient outcomes in pancreatic cancer. Positive results would elevate ctDNA from a prognostic biomarker to a predictive biomarker that guides treatment decisions.
Key Ongoing Trials:
- ESPAC-6 Trial: ctDNA-guided adjuvant therapy in resectable PDAC - randomized trial testing whether ctDNA-directed treatment intensification prevents recurrence in high-risk patients
- CONKO-009 Trial: MRD-directed treatment strategies - evaluating ctDNA-guided escalation/de-escalation of adjuvant therapy to improve outcomes while minimizing toxicity
- Multiple KRAS G12D inhibitor trials: Testing novel G12D-specific inhibitors (MRTX1133 and others) in the ~40% of patients with this mutation variant
- HER2-targeted therapy trials: Evaluating trastuzumab deruxtecan and other HER2-directed agents in the ~2% of patients with HER2 amplification
Clinical Significance: These trials aim to provide Level 1 evidence (randomized controlled trials) demonstrating that acting on ctDNA results improves survival, not just prognostic stratification. Positive results from ESPAC-6 or CONKO-009 would fundamentally change the role of ctDNA in pancreatic cancer management from research tool to standard-of-care decision-making instrument.
Clinical Summary and Recommendations
ctDNA testing in pancreatic cancer addresses critical unmet needs in a disease with poor prognosis and limited treatment options, but must be applied with full understanding of its biological limitations.
Evidence-Based Clinical Recommendations
MRD Monitoring (Prognostic Role):
- Strongest Evidence: Post-resection ctDNA detection predicts recurrence risk (HR 3.9-12.2, depending on study)
- Lead Time Advantage: 4.5-6.5 months earlier detection than imaging
- Superior to CA19-9: ctDNA more strongly associated with recurrence (HR 7.8 vs 1.9)
- Critical Limitation: Sensitivity of 43-71% means negative ctDNA does NOT exclude residual disease
- Best Practice: Longitudinal monitoring (serial timepoints) improves detection over single assessment
- Current Status: Provides risk stratification; interventional trials (ESPAC-6, CONKO-009) testing treatment modification strategies ongoing
Molecular Profiling (Actionable Genotyping):
- KRAS G12C (1-2%): Sotorasib available (ORR 26.7%); identify for approved targeted therapy
- KRAS G12D (~40%): Identify for clinical trial enrollment; emerging inhibitors (MRTX1133) in development - clinically significant for largest molecular subgroup
- BRCA1/2 (4-7% germline, additional somatic): Olaparib maintenance therapy (Level 1 evidence: PFS 7.4 vs 3.8 months, HR 0.53)
- MSI-H/dMMR (1-2%): Pembrolizumab immunotherapy (ORR ~30%)
- NTRK fusions (<1%): Larotrectinib/entrectinib (ORR ~75%)
- HER2 amplification (~2%): Clinical trials of trastuzumab deruxtecan
- Clinical Advantage: Non-invasive alternative when pancreatic biopsy is high-risk or tissue insufficient
Appropriate Use Scenarios:
- Post-resection risk stratification: Identify high-risk patients for closer surveillance or clinical trial enrollment
- Metastatic disease genotyping: When tissue biopsy is not feasible or tissue is exhausted from prior testing
- Treatment response monitoring: Serial ctDNA during systemic therapy may predict clinical outcomes
- Clinical trial eligibility: Molecular profiling for KRAS G12D, HER2, or other biomarker-selected trials
Scenarios Where ctDNA Has Limited Value:
- Standard post-resection surveillance when not participating in clinical trials (CA19-9 + imaging remain standard)
- When negative result would not change management (due to low sensitivity and false-negative risk)
- Very early-stage disease where low tumor burden limits detection
- When tissue biopsy is readily available and provides equivalent information
Bottom Line: Pancreatic cancer demonstrates both the promise and limitations of ctDNA testing. While MRD detection provides strong prognostic information (HR 3.9-12.2) with substantial lead time (4.5-6.5 months), the "low-shedder" biology results in sensitivity of only 43-71%, limiting its negative predictive value. Molecular profiling identifies actionable targets in a meaningful subset of patients - particularly emerging KRAS G12D inhibitors for ~40% of patients and proven PARP inhibitor therapy for BRCA-mutated disease. However, most pancreatic cancer patients lack targetable alterations, and ctDNA cannot replace standard monitoring with CA19-9 and imaging. The field awaits results from interventional trials (ESPAC-6, CONKO-009) to establish whether acting on ctDNA results improves survival outcomes.
References
- Groot VP, Mosier S, Javed AA, et al. Circulating tumor DNA as a clinical test in resected pancreatic cancer. Clin Cancer Res. 2019;25:4973-4984.
- 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.
- Watanabe F, Suzuki K, Tamaki S, et al. Longitudinal monitoring of circulating tumor DNA in resectable pancreatic cancer. Ann Surg Oncol. 2019;26:197-204.
- Strickler JH, Satake H, George TJ, et al. Sotorasib in KRAS p.G12C-mutated advanced pancreatic cancer. N Engl J Med. 2023;388:33-43.
- Bekaii-Saab TS, Yaeger R, Spira AI, et al. Adagrasib in advanced solid tumors harboring a KRAS G12C mutation. J Clin Oncol. 2023;41:4097-4106.
- 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.
Evidence summary as of January 2026 | Document Version: 2.0