A Practical Guide to Biomarkers in Pancreatic Cancer

Biomarkers turn the story of pancreatic cancer into something we can measure, track, and use to steer care. If imaging is the map, biomarkers are the mile markers and fuel gauge. They won't replace clinical judgment or a high quality scan, but they can sharpen decisions at every step — from assessing risk to monitoring treatment response. This guide translates the core biomarkers used in pancreatic cancer into plain language, with honest caveats and a focus on what's useful right now.

What counts as a biomarker in pancreatic cancer?

A biomarker is any measurable signal from your body that helps with diagnosis, prognosis, or treatment selection. In pancreatic cancer, biomarkers fall into a few common categories:

• Blood proteins measured by lab tests (for example, CA 19-9)
• Tumor DNA changes found in biopsy tissue (such as KRAS or BRCA alterations)
• Inherited DNA changes that raise risk (germline variants like BRCA2 or PALB2)
• Fragments of tumor DNA in blood (circulating tumor DNA, or ctDNA)
• Cyst fluid markers when a pancreatic cyst is being evaluated (CEA, glucose, and select gene mutations)

Each category answers a different question. No single test tells the whole story.

How biomarkers support the care journey

1) Risk and early detection

Today, there is no validated blood test to screen the general population for pancreatic cancer. The most studied blood marker, CA 19-9, is not specific enough for screening. Major guidelines recommend surveillance only for clearly high risk individuals — those with certain inherited syndromes (like Peutz-Jeghers, CDKN2A, BRCA2, PALB2, or STK11), or a strong family history. Surveillance in these groups usually relies on MRI/MRCP and endoscopic ultrasound at specialized centers. New-onset diabetes after age 50 can be an early clue in some people, but it is not a diagnosis by itself; most new-onset diabetes is not cancer related.

Research is active on multi-marker panels, metabolomics, and ctDNA for early detection, but these are not yet routine care. If you've seen headlines about "blood tests that find cancer," know that evidence in pancreatic cancer is still developing.

2) Diagnosis and staging

Once a pancreatic mass is identified on imaging, biomarkers play a supporting role. CA 19-9 is often measured at baseline alongside a biopsy and pancreatic protocol CT or MRI. CA 19-9 can suggest tumor burden and help with prognosis in context, but imaging and tissue diagnosis carry the weight for staging and treatment planning.

3) Treatment selection

This is where tumor and inherited DNA testing matter. Every person diagnosed with pancreatic adenocarcinoma should be offered germline (inherited) genetic testing,¹ and tumor profiling is increasingly standard. These tests can reveal:

• DNA repair defects that predict sensitivity to platinum chemotherapy²
• Rare but actionable targets like NTRK fusions, BRAF V600E, HER2 amplification, or KRAS G12C
• Microsatellite instability or mismatch repair deficiency, which can predict benefit from immunotherapy in a small subset³

Even when results don't point to a targeted drug, they can inform prognosis and clinical trial options.

4) Monitoring and survivorship

Once treatment begins, biomarkers become a dashboard. Trends in CA 19-9 can move earlier than scans, offering an early signal of response or progression. ctDNA is an emerging tool for tracking minimal residual disease after surgery or detecting recurrence earlier than imaging,⁴ though it's not yet a universal standard. After curative-intent treatment, many teams follow a combination of symptoms, labs, and periodic imaging rather than a single test.

Blood biomarkers you'll hear about

CA 19-9: the workhorse with important limits

What it is: CA 19-9 is a carbohydrate antigen shed by many pancreatic tumors into the bloodstream. Most labs set the upper limit of normal around 37 U/mL, but ranges can vary.

How it's used:^{5 6}

• Baseline measurement before treatment to help with prognosis in context
• Monitoring during chemotherapy or after surgery to assess response
• Correlating with imaging over time to track the trajectory of disease

What it is not: a screening test for the general population. CA 19-9 can be elevated by non-cancer causes like bile duct obstruction, pancreatitis, cholangitis, and even poorly controlled diabetes. It can also be normal in people who have cancer, especially if they do not make the "Lewis antigen," a genetic trait required to produce CA 19-9. About 5 to 10 percent of people fall into this category, and the frequency varies by ancestry.

Key pitfalls and how teams navigate them:

• Biliary obstruction can push CA 19-9 very high. Levels often fall after a stent is placed and bile flows again. Many clinicians recheck CA 19-9 1 to 3 weeks after decompression to get a cleaner baseline.
• Assay variability exists across labs. If possible, follow trends using the same laboratory method over time.
• Biotin supplements can interfere with certain immunoassays. If your lab uses a biotin-streptavidin method, high-dose biotin may skew results. Disclosing supplement use helps the lab and care team interpret the number.
• Inflammation from pancreatitis or infection can transiently elevate CA 19-9. Context and repeat testing matter.

How to think about the number: a single CA 19-9 value is a snapshot; the trend is the movie. If the level drops steadily with treatment and imaging shows shrinkage, that's a coherent story. If the level rises but scans are stable, your team may repeat the test, check for obstruction, and correlate with symptoms before changing course. It's a bit like watching your credit score after paying down a loan — one data point is less meaningful than the direction over several months.

CEA and CA-125: supporting actors

CEA (carcinoembryonic antigen) and CA-125 can be elevated in some cases of pancreatic cancer but they are less sensitive and less specific than CA 19-9. They are sometimes followed when a patient's tumor does not secrete CA 19-9 or when these markers were elevated at baseline. Elevated CEA can also come from conditions like smoking, inflammatory bowel disease, and other cancers. These markers are adjuncts, not stand-alone decision makers.

Inflammation markers and composite scores

Ratios like neutrophil-to-lymphocyte ratio and markers like CRP are being studied for prognosis. They reflect the body's systemic inflammatory response rather than tumor biology directly. Some studies link higher inflammation to poorer outcomes, though these markers are not routinely used to guide therapy in pancreatic cancer. They may provide context alongside primary tests.

Tissue and genomic biomarkers from the tumor

Common drivers you'll see on a tumor report

Pancreatic ductal adenocarcinoma (PDAC) often carries a core set of genetic alterations: KRAS mutations in about 90 percent, plus changes in TP53, CDKN2A, and SMAD4. These shape tumor behavior but usually don't point to approved targeted therapies. Still, knowing the profile can help identify clinical trials and refine prognosis.

Actionable or potentially actionable findings

• MSI-high or mismatch repair deficiency: Rare in PDAC, but when present, it predicts benefit from immune checkpoint blockade based on robust tissue-agnostic data.³
• NTRK fusions: Very rare, but targeted TRK inhibitors have durable responses across cancers.
• BRAF V600E: Uncommon in PDAC. When present, BRAF-targeted combinations are considered in line with tissue-agnostic guidance.
• HER2 amplification: Seen in a small minority. HER2-directed strategies may be considered where supported by evidence.
• KRAS G12C: A small subset carries this specific KRAS variant. KRAS G12C inhibitors have shown activity in early studies; access and indications continue to evolve.

A credibility note: these alterations are uncommon in PDAC. Most patients will not have a directly targetable mutation, which is why chemotherapy remains the backbone of treatment. When a rare target is found, tumor boards often review the evidence and consider clinical trials when available.

DNA damage repair (DDR) and homologous recombination deficiency (HRD)

Alterations in BRCA1, BRCA2, PALB2, and related genes impair the cell's ability to fix double-strand DNA breaks. In pancreatic cancer, this biology can translate into increased sensitivity to platinum chemotherapy.^{2 7} PARP inhibitors have been studied in selected patients with BRCA or PALB2 alterations. The details of regulatory indications and best use are evolving, and guidelines are periodically updated. The consistent takeaway is that identifying DDR alterations helps tailor chemotherapy choices and opens doors to clinical trials focused on HRD biology.

Transcriptional subtypes and GATA6

RNA-based profiling can classify PDAC into "classical" and "basal-like" subtypes. The classical subtype, often marked by higher GATA6 expression, is associated with a more differentiated biology, while basal-like tumors behave more aggressively. Early data suggest subtype may correlate with response to certain chemotherapy backbones, but routine use of RNA subtyping is still maturing and varies by center.

PD-L1 expression

PD-L1 staining is sometimes reported, but in PDAC it has not been a reliable predictor of immunotherapy response outside the MSI-high context. It is one data point among many and should not be overinterpreted.

Pharmacogenomics that influence chemotherapy tolerance

• DPYD variants can raise the risk of toxicity from 5-FU or capecitabine. Some centers screen for DPYD to calibrate dosing.
• UGT1A1 variants, especially *28/*28, increase risk of neutropenia with irinotecan. Knowing the genotype can guide safer dosing strategies.

These tests don't change whether a drug "works" against the cancer, but they help tailor dosing to reduce avoidable side effects.

Inherited (germline) testing: it's for everyone with PDAC

Major guidelines recommend germline testing for all individuals diagnosed with pancreatic adenocarcinoma, regardless of family history.¹ Why? Because actionable findings are missed if testing is limited only to those with visible risk factors. Germline results can:

• Inform treatment choices, particularly around platinum sensitivity and HRD biology
• Identify MSI-associated syndromes that may change management
• Guide surveillance for other cancers
• Enable cascade testing for relatives who might benefit from preventive care

Genes commonly included: BRCA1, BRCA2, PALB2, ATM, CDKN2A, STK11, PRSS1, MLH1, MSH2, MSH6, PMS2, and EPCAM, among others. Pre-test and post-test counseling clarifies what a positive, negative, or uncertain result means. For families, this is not just about pancreatic cancer risk; several of these genes influence breast, ovarian, colorectal, melanoma, and other cancer risks as well.

Liquid biopsy and ctDNA: the up-and-coming tool

Circulating tumor DNA testing captures fragments of tumor DNA shed into the blood. In pancreatic cancer, ctDNA can detect tumor-specific mutations like KRAS and track them over time.^{8 9} Strengths and limits:

• Sensitivity depends on tumor burden and location. Small or localized tumors may not shed enough DNA to be detected.
• When detected, ctDNA levels can change before imaging, offering an early hint of response or progression.
• After surgery, persistent or rising ctDNA may signal minimal residual disease, although how best to act on that information is still under study.⁴
• ctDNA does not replace tissue for comprehensive profiling. If a targetable mutation is suspected, tissue testing remains the gold standard when feasible.

In short, ctDNA is a promising complement to existing tools. It adds value in many cases, but a negative result does not rule out disease.

Biomarkers in pancreatic cysts: a different question

Many people discover a pancreatic cyst incidentally on a scan. Most cysts are benign, some are premalignant, and a minority carry or harbor risk of cancer over time. When evaluating a cyst, the biomarker toolkit shifts:¹⁰

• Cyst fluid CEA helps distinguish mucinous from non-mucinous cysts
• Cyst fluid glucose is low in mucinous cysts and can be a practical marker
• KRAS and GNAS mutations support a diagnosis of intraductal papillary mucinous neoplasm (IPMN)
• VHL mutations point toward serous cystadenoma

These tests support risk stratification alongside imaging details like duct size and mural nodules. They are not the same as biomarkers used once pancreatic cancer is established.

How labs and assays can shape results

Numbers come from methods, and methods matter. A few practical realities help with interpretation:

• Biological factors: Lewis antigen status determines whether a person can produce CA 19-9 at all. Cholestasis, pancreatitis, and uncontrolled diabetes can elevate levels from non-cancer causes.

These aren't footnotes — they are the guardrails that prevent misinterpretation.

Putting numbers in context: practical examples

Here are realistic scenarios that illustrate how clinicians think through biomarker data:

• A person presents with jaundice, a pancreatic head mass, and CA 19-9 of 2,500 U/mL. A biliary stent is placed, bilirubin falls, and CA 19-9 drops to 450 U/mL over two weeks. Chemotherapy starts, the next scan shows shrinkage, and CA 19-9 declines to 80 U/mL. The trend matches the imaging story — despite the initially alarming number.
• Another patient has measurable metastatic disease on imaging but a normal CA 19-9. Lewis antigen testing reveals a genotype consistent with non secretion. For monitoring, the team relies on imaging and symptoms, and may track an alternative marker like CEA if elevated at baseline.
• After surgical resection, CA 19-9 falls into the normal range and stays there for months. Later, it rises from 18 to 62 U/mL without symptoms. A repeat test confirms the increase and imaging finds a small recurrence. Here, the marker provided an early nudge to look closer.
• ctDNA detects KRAS G12D at baseline. After two chemotherapy cycles, the variant allele fraction falls by 80 percent. Although not a substitute for a scan, this kind of shift can complement radiographic response and support staying the course.

In every scenario, decisions are made by triangulating biomarkers, imaging, and clinical context. No single test is in the driver's seat.

What about sex, age, and ancestry differences?

Unlike some cancers, standard biomarker cutoffs in pancreatic cancer do not routinely differ by sex or age. One relevant difference by ancestry is the prevalence of Lewis antigen non secretion, which disables CA 19-9 production. This trait occurs in a notable minority across populations and can be more common in certain groups. If CA 19-9 is unexpectedly low or unhelpful, discussing Lewis antigen status can clarify whether the marker is usable for that individual.

Where evidence is strong — and where it's still growing

Strong consensus areas:

• CA 19-9 is useful for baseline assessment and monitoring, not for screening the general population.^{5 6}
• All patients with pancreatic adenocarcinoma should be offered germline genetic testing.
• Tumor profiling can uncover rare but meaningful targets and should be considered when feasible.
• MSI-high and NTRK fusion positive tumors are candidates for tissue-agnostic therapies with supportive data.

Areas under active study:

• ctDNA to guide adjuvant therapy decisions and detect minimal residual disease
• RNA subtyping to select chemotherapy backbones
• Early detection panels combining proteins, DNA, and metabolites
• How best to integrate HRD signatures beyond single gene testing

As always, guideline updates follow the data, and multidisciplinary tumor boards help adapt new evidence responsibly.

Questions that help you get the most from biomarker testing

• Which biomarkers are being used to track my specific cancer, and why were they chosen?
• If CA 19-9 is not informative for me, what alternatives are we following?
• Has my tumor been profiled for targetable alterations and DNA repair defects?
• Was germline testing performed, and what do the results mean for my care and family members?
• Are the same assay methods being used for my serial tests so we can trust trends?
• Could ctDNA add value in my situation, and how would results change management?

Responsible caveats to keep in mind

• A biomarker trend can be misleading if bile ducts are blocked, infection is present, or lab methods change between draws.
• Normal biomarkers do not rule out cancer progression. Imaging and symptoms remain central.
• Actionable targets are rare in PDAC. When they appear, clinical trials are often the best route to access emerging therapies.
• New tests are exciting, but adopting them before evidence matures can add cost and confusion without improving outcomes.

Key takeaways

• Use biomarkers as tools, not verdicts. They work best when paired with high quality imaging and clinical context.
• CA 19-9 is valuable for monitoring but limited by biology and lab factors. Trend it, don't worship it.
• Germline testing for all and tumor profiling when feasible can uncover DNA repair defects and rare targets that influence therapy.
• ctDNA is emerging as a sensitive companion for tracking disease, particularly after surgery, though its routine use is still evolving.
• When a biomarker result seems out of sync with the clinical picture, look for reversible causes like biliary obstruction or assay issues and consider repeating the test.

Bottom line: biomarkers help translate a complex disease into measurable signals that can guide thoughtful care. They're powerful when used responsibly, with a clear understanding of what they can and cannot tell us. In pancreatic cancer, that balance — grounded in evidence and personalized to the individual — is what turns numbers into meaningful decisions.