KRAS mutation awareness: what precision oncology patients search
KRAS mutations appear in roughly 25% of all human cancers, yet most precision oncology platforms treat patient search behavior around KRAS as background noise. KRAS mutation data, when mapped against behavioral signals, reveals where patients are in their treatment arc and what they will do next. This post covers what precision oncology patients actually search, what those signals mean for pharma and clinical trial teams, and why the data layer underneath matters more than the search itself.
Approximately 25% of all human cancers carry a KRAS mutation. For decades, KRAS was considered undruggable. That changed in 2021 when the FDA approved sotorasib for KRAS G12C-mutated non-small cell lung cancer. Now the search behavior of patients navigating KRAS-driven cancers has become one of the most information-dense behavioral datasets in precision oncology.
The top-ranking content on this topic focuses on clinical mechanisms: which KRAS variants appear in which tumor types, how inhibitors bind, what trial data looks like. None of it addresses the behavioral layer. None of it asks what patients themselves are searching, when they search it, or what those patterns predict about clinical decisions downstream.
That gap matters. KRAS mutation data is not just a molecular finding. It is a behavioral trigger. The moment a patient receives a KRAS-positive result, their search behavior shifts in ways that are measurable, predictable, and clinically meaningful.
What cancer is KRAS associated with
KRAS is an oncogene. When mutated, it produces a protein that is permanently switched on, driving uncontrolled cell growth. KRAS mutations are found across multiple solid tumor types, but three cancers dominate the distribution.
Pancreatic ductal adenocarcinoma carries KRAS mutations in approximately 90% of cases. Colorectal cancer carries them in roughly 40% of cases. Non-small cell lung cancer (NSCLC) carries them in about 25-30% of cases. These three tumor types account for the vast majority of KRAS-mutated cancers seen in clinical practice.
Smaller but significant populations of KRAS mutations appear in endometrial cancer, cholangiocarcinoma, and certain subtypes of ovarian cancer. The KRAS gene itself sits on chromosome 12 and encodes a GTPase protein that acts as a molecular switch in cell signaling pathways, specifically the RAS/MAPK pathway.
When patients search "KRAS mutation," they are overwhelmingly arriving from one of these three cancer types. VIOLET behavioral mapping shows that pancreatic cancer patients search KRAS within days of diagnosis, colorectal cancer patients search it after learning their tumor is RAS-mutated (often in the context of anti-EGFR therapy eligibility), and lung cancer patients search it after molecular profiling returns.
What percentage of cancers have KRAS mutations
Across all human cancers, KRAS mutations appear in approximately 25% of cases. That makes KRAS the single most commonly mutated oncogene in human cancer.
The specific variant matters. KRAS G12D is the most frequent mutation overall, appearing in roughly 36% of all KRAS-mutated tumors. KRAS G12V accounts for about 23%. KRAS G12C, the first variant to be successfully drugged, accounts for approximately 14% of all KRAS mutations but is disproportionately common in NSCLC, where it represents about 13% of all cases.
This distribution creates a behavioral signal. Patients searching "KRAS G12C" are almost always lung cancer patients who have received molecular profiling results. Patients searching "KRAS G12D" are more likely pancreatic cancer patients. The mutation subtype in the search query itself is a proxy for tumor type, disease stage, and likely treatment pathway.
What drug is targeting KRAS
Sotorasib (Lumakras, developed by Amgen) was the first FDA-approved KRAS inhibitor, receiving approval in May 2021 for KRAS G12C-mutated NSCLC after prior systemic therapy. Adagrasib (Krazati, developed by Mirati Therapeutics, now part of Bristol Myers Squibb) followed with FDA approval in December 2022 for the same indication.
As MedPage Today's expert panel discussed in June 2021, the approval of sotorasib represented a fundamental shift in how clinicians think about KRAS. A target that was considered undruggable for over 30 years suddenly had a small molecule inhibitor in clinical use.
The pipeline has expanded rapidly since then. KRAS G12D inhibitors are now in clinical trials, which matters enormously because G12D is the most common KRAS variant. KRAS G12V inhibitors and pan-KRAS inhibitors are also in development. Drugs targeting KRAS degradation (rather than direct inhibition) represent another active area.
Behavioral data shows a clear search pattern around these drugs. Within 72 hours of the sotorasib approval announcement, search volume for "KRAS G12C treatment" increased over 400%. Similar spikes followed the adagrasib approval. These are not casual searches. They correlate with downstream actions: clinical trial queries, second opinion requests, and treatment center searches.
What is the biggest breakthrough in pancreatic cancer
Pancreatic cancer remains one of the most lethal malignancies, with a five-year survival rate of approximately 12%. The biggest breakthrough is not a single drug but the convergence of three developments: early detection biomarker research, KRAS-targeted therapy development, and improved combination regimens.
For KRAS specifically, the development of G12D inhibitors is the most consequential advance for pancreatic cancer patients because roughly 40% of KRAS-mutated pancreatic cancers carry the G12D variant. Early-phase clinical trial data from MRTX1133 and other G12D-targeting compounds have generated significant patient interest.
As noted in MedPage Today's coverage of early pancreatic cancer detection, liquid biopsy approaches that detect circulating tumor DNA with KRAS mutations may eventually enable earlier diagnosis. This is critical because most pancreatic cancers are diagnosed at stage III or IV.
Behavioral signals around pancreatic cancer and KRAS are distinct from other tumor types. Pancreatic cancer patients and their caregivers search with higher urgency, more frequently at night, and with more treatment-specific queries earlier in the disease course. This pattern aligns with what we have documented in pancreatic cancer search behavior and early detection signals.
Key statistics
KRAS mutation data and precision oncology behavioral signals produce specific, quantifiable patterns that pharma teams, trial sponsors, and health systems should track.
The behavioral signal map for KRAS-mutated patients
VIOLET tracks over 750 oncology-related search terms and maps them to behavioral phases. For KRAS-mutated patients, the behavioral arc follows a consistent pattern across tumor types, with timing variations.
Phase one is molecular profiling awareness. Patients search terms like "what is molecular profiling," "tumor DNA testing," and "biomarker testing for cancer." This phase occurs between diagnosis and the return of profiling results, typically a 7-21 day window.
Phase two is mutation-specific education. Once results arrive, searches shift to "KRAS mutation meaning," "KRAS G12C prognosis," "is KRAS mutation bad," and "KRAS mutation life expectancy." This phase peaks within 48 hours of result delivery and remains elevated for about two weeks.
Phase three is treatment matching. Patients search for specific drugs by name, clinical trial identifiers, and treatment center specializations. Queries like "sotorasib side effects," "KRAS G12C clinical trials near me," and "best hospital for KRAS lung cancer" define this phase.
Phase four is second opinion and validation. Patients who have already been presented with a treatment plan search for confirmation or alternatives. This phase overlaps heavily with the patterns documented in oncology second opinion seeking behavior.
Why current KRAS content fails patients
The content currently ranking for KRAS-related queries is written for oncologists and researchers. It assumes familiarity with GTPase biochemistry, RAS signaling cascades, and clinical trial design. Patients searching "KRAS mutation" do not have that background.
The top three results in the SERP are peer-reviewed publications or clinical news summaries. They answer the question "what does KRAS do molecularly" but not the question patients are actually asking: "what does my KRAS result mean for my treatment options right now."
This content gap creates a behavioral signature. Patients who land on technical KRAS content bounce quickly and then refine their search with simpler language. The refinement pattern itself is a signal. A patient who searches "KRAS mutation colorectal cancer" and then searches "can I still get Erbitux with KRAS mutation" within the same session has just revealed their exact clinical scenario and treatment concern.
Precision oncology behavioral signals as intelligence
Pharma companies spend hundreds of millions on market research to understand where patients are in their treatment decision arc. Precision oncology behavioral signals provide that intelligence at scale, in real time, without surveys or focus groups.
Consider what a single search session reveals. A patient who searches "KRAS G12C sotorasib vs adagrasib" is comparing two approved therapies. They are past the education phase. They are in active treatment selection. A patient who searches "KRAS G12D clinical trials phase 2" is looking for options that do not yet exist in the approved setting. They are likely a pancreatic cancer patient whose current treatment is failing or has failed.
These signals have direct commercial value. A pharma company launching a KRAS G12D inhibitor can use behavioral intelligence to identify the geographic concentration of patients searching for G12D-specific options, the timing of their search intensity relative to treatment failure, and the specific comparator questions they ask. That intelligence shapes launch sequencing, KOL engagement, clinical trial site selection, and HCP education priorities.
We have covered how pharma teams can operationalize this kind of signal in what pharma companies can learn from cancer search data before campaigns launch.
The data trust problem underneath KRAS mutation data
Molecular profiling results, including KRAS mutation status, sit at the intersection of genomic data, laboratory data, and clinical data. Each of those data types has a different provenance chain, a different consent profile, and a different recency characteristic.
A KRAS mutation result from a Foundation Medicine report has different provenance than one from a Tempus xT panel, which has different provenance than one from an academic medical center's in-house next-generation sequencing lab. If an AI model trains on KRAS mutation data without scoring the provenance, consent, and quality of each data point, the model inherits every inconsistency in the underlying data.
This is where the Data Trust Index applies directly. DTI scores every health data record across eight dimensions: Provenance (25%), Consent (20%), Recency (15%), Quality (10%), Concordance (10%), Validation (10%), Breadth (5%), and Stability (5%). For KRAS mutation data specifically, provenance and concordance are the critical dimensions. Does the KRAS result match across the lab report, the EHR problem list, and the treatment record? If not, the DTI score drops, and any downstream model should know that.
Our work with imaware demonstrated this at scale. Across 105,000 diagnostic records, standardization and trust scoring reduced processing time from three weeks to two hours while surfacing discordances that manual review had missed. The same approach applies to molecular oncology data. As we have detailed in genomic data trust: provenance requirements for precision medicine, genomic results require the highest provenance standards because treatment decisions depend on single-variant accuracy.
The clinical trial recruitment signal
KRAS G12C inhibitors were approved based on relatively small single-arm trials. Sotorasib's CodeBreaK 100 enrolled 124 patients. Adagrasib's KRYSTAL-1 enrolled 116 patients. Recruiting patients with specific KRAS variants remains a bottleneck.
Behavioral intelligence identifies patients who are actively searching for trial options before they contact a site. A patient who searches "KRAS G12C clinical trial enrolling now" is a self-identified candidate. A patient who searches "KRAS mutation treatment options after chemo failed" is a candidate who does not yet know specific trials exist.
The difference between those two searches is the difference between a patient who needs a site referral and a patient who needs education first. Trial sponsors who can distinguish between these populations can allocate recruitment resources more efficiently. We have covered the broader recruitment challenge in behavioral intelligence for clinical trial recruitment in oncology.
What happens at 2 AM
KRAS-related searches follow the same nocturnal pattern we have documented across oncology behavioral data. Search volume for KRAS mutation terms peaks between 10 PM and 2 AM, when patients are alone with their diagnosis and their questions.
The 2 AM search window is significant because it correlates with treatment decision inflection points. Patients searching late at night are processing information they received during the day. They are reading clinical trial descriptions, comparing survival statistics, and formulating questions for their next oncology visit.
As we have described in how the 2 AM search window predicts clinical trial enrollment 90 days out, nocturnal search intensity in the weeks following a molecular profiling result is one of the strongest predictors of whether a patient will seek a second opinion or enroll in a clinical trial within the next quarter.
The expanding druggable landscape changes patient behavior
The KRAS inhibitor pipeline is no longer limited to G12C. As researchers work to expand precision oncology by drugging all KRAS mutants, patient search behavior is evolving in parallel. Three years ago, a patient with a KRAS G12D mutation had no reason to search for targeted therapy options because none existed. Now those patients search actively because early-phase trial results have reached the news cycle.
This creates a temporal signal. The moment a new KRAS inhibitor enters Phase 1 trials, search volume for that specific variant increases within weeks. When Phase 2 data is presented at ASCO or AACR, another spike follows. Each data readout generates a measurable behavioral response that can be tracked, quantified, and used to predict patient flow.
For pharma teams, this means targeted therapy patient intelligence is not static. The behavioral landscape around KRAS mutations changes with every clinical milestone. An intelligence platform that scored KRAS behavioral data in 2022 would produce different signals than one scoring the same queries in 2025, because the actionable treatment landscape has shifted.
Why the trust layer matters more than the signal
Behavioral signals are only useful if the data underneath them is trustworthy. A search pattern that appears to show increasing interest in KRAS G12D inhibitors could reflect genuine patient need, or it could reflect content marketing campaigns by trial sponsors. Without provenance scoring, you cannot distinguish signal from noise.
Similarly, KRAS mutation data in clinical datasets is only as reliable as the molecular profiling report it came from, the EHR system that recorded it, and the data pipeline that transmitted it. A KRAS G12C result that was entered manually into an EHR as free text has lower reliability than one that flowed directly from a CLIA-certified lab via structured HL7 messaging.
The Data Trust Index makes this distinction explicit. Every KRAS mutation data point can be scored for how it was generated (Provenance), whether the patient consented to its use in research or AI training (Consent), how recently it was validated (Recency), and whether it matches across multiple data sources (Concordance). That scored data is what makes precision oncology behavioral signals actionable rather than anecdotal.
VIOLET maps behavioral signals across 750+ oncology search terms before patients reach a clinic. If your team is working on cohort identification, trial recruitment, or oncology market intelligence for KRAS-mutated patient populations, schedule a conversation with the SuperTruth commercial team or (215) 918-4140.
Further reading:
Jason Alan Snyder
Co-founder of SuperTruth and Artists & Robots, and an inventor on the Data Trust Index patents. Twenty-plus years building technology inside Interpublic Group. He writes here nearly every day on data trust, provenance, and what AI should be allowed to act on, and publishes essays on his Substack.
About SuperTruth · LinkedIn · Substack · jasonalansnyder.com
See it in practice
750+ cancer search terms. Live in production.
VIOLET maps behavioral signals 12 to 18 months before clinical presentation.