Lung Cancer Pills That Target EGFR May Quietly Weaken the Heart, Study Warns

Third-generation EGFR tyrosine kinase inhibitors have transformed the outlook for patients whose lung cancers carry mutations in the epidermal growth factor receptor, turning a disease that once carried a grim prognosis into a chronic condition that can often be managed for years with a daily pill. Osimertinib, the best-known drug in this class, has become a global standard of care, and newer agents such as furmonertinib, developed and widely used in China, have followed in its footsteps. Yet as survival curves stretch outward, oncologists are increasingly confronting an uncomfortable question: what price do the heart and blood vessels pay during years of continuous targeted therapy? A new retrospective study from Chinese hospitals, published in BMC Cancer, adds a sobering data point to that conversation, documenting serious cardiovascular toxicity in a small group of patients treated with two different third-generation EGFR inhibitors.

The research team, led by pharmacists at the First Affiliated Hospital with Nanjing Medical University together with colleagues in Sichuan, set out to answer a question that earlier trials largely left unaddressed. Cardiovascular side effects had been reported during osimertinib treatment before, including reductions in the heart’s pumping efficiency and dangerous changes in the heart’s electrical rhythm, but evidence for other drugs in the same class remained thin. Because several third-generation EGFR inhibitors now share clinical territory, doctors needed to know whether these cardiac problems were a quirk of one molecule or a broader liability of the class. The researchers therefore screened consecutive patients with non-small cell lung cancer who had received third-generation EGFR tyrosine kinase inhibitor therapy at their center between January 2020 and June 2025, building a real-world picture rather than relying on the carefully filtered populations of registration trials.

The screening process illustrates the practical difficulty of studying cardiac toxicity in oncology. Of 745 patients treated with third-generation EGFR inhibitors during the study window, only 26 had complete paired baseline and follow-up cardiac assessments that could be compared side by side. That attrition reflects a wider gap in clinical practice: cardiac monitoring before and during targeted cancer therapy is not standardized, and many patients begin treatment without a documented echocardiogram or electrocardiogram that could serve as a reference point. The small analyzable sample is a limitation the authors openly acknowledge, and it shapes how far the findings can be pushed. Still, the data that did emerge offer concrete technical detail about what happens to the heart when these drugs are administered.

The headline number is striking in its rarity. Among the 745 screened patients, cancer therapy-related cardiovascular toxicity, abbreviated CTR-CVT in the cardio-oncology literature, was documented in just 8 patients, an incidence of roughly 1.1 percent during the study period. Every case was graded as Grade 3 under version 5.0 of the Common Terminology Criteria for Adverse Events, the standardized scale oncologists use to classify the severity of side effects. Grade 3 is not a mild finding; it denotes severe toxicity that typically demands intervention, dose modification, or interruption of the cancer treatment itself. In other words, when cardiovascular toxicity appeared in this cohort, it was not a subtle laboratory abnormality but a clinically consequential event.

The phenotypes of those events are equally informative. Five patients developed heart failure with reduced ejection fraction, the condition in which the left ventricle, the heart’s main pumping chamber, loses the contractile strength needed to circulate blood effectively. Seven patients developed prolongation of the corrected QT interval, or QTc, an electrocardiographic measure of the time the ventricles take to depolarize and repolarize with each heartbeat. Excessive QTc prolongation is dangerous because it can degenerate into torsades de pointes, a polymorphic ventricular tachycardia that can cause sudden cardiac arrest. Four patients experienced both problems at once, a combination that is particularly challenging to manage because the drugs used to treat one problem can aggravate the other. Importantly, the toxicity was not confined to a single agent: it occurred in 3 of 17 patients receiving osimertinib and, notably, in 5 of 5 patients receiving furmonertinib, a distribution that raises questions about whether the risk extends across the class.

To understand the mechanism underlying these numbers, the researchers turned to serial cardiac functional parameters in the 26 patients with complete assessments. Left ventricular ejection fraction, the standard echocardiographic index of pumping function, fell significantly from an average of 63.2 percent at baseline to 57.0 percent at follow-up, a decline that reached statistical significance with a P value of 0.014. The left ventricular end-diastolic diameter, which measures the chamber’s internal dimension when it is fully relaxed and filled, increased significantly over the same interval, with a P value of 0.048, a pattern consistent with ventricular dilation under stress. When the team stratified patients by toxicity status, the differences sharpened: the CTR-CVT group showed significant intra-group declines in ejection fraction and significant prolongation of the QTc interval, both with P values below 0.01, while the non-CTR-CVT group showed no significant changes in either parameter. That contrast suggests the toxicity events were not random noise but reflected a genuine, measurable drift in cardiac structure and electrical function in susceptible patients.

The clinical trajectory of these patients offers a measure of reassurance. Most achieved cardiac improvement during follow-up, and those who developed heart failure received standardized pharmacotherapy, the toolkit of modern heart failure management that includes angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, angiotensin receptor-neprilysin inhibitors, sodium-glucose co-transporter 2 inhibitors, and beta-blockers. The fact that the cardiac function of most patients recovered indicates that vigilance and prompt intervention can catch and reverse the damage, provided the decline is detected early enough. That, in turn, underscores why the absence of baseline and serial cardiac assessments in most of the 745 screened patients is itself a finding: a toxicity that is invisible cannot be treated until it announces itself through symptoms.

The authors are careful, and appropriately so, about what their data cannot prove. In their conclusions they state that the small sample size, non-standardized cardiac surveillance, frequent concomitant systemic therapy, and residual confounding preclude causal attribution to individual agents or confirmation of a class effect. This is a crucial caveat. Many patients with advanced EGFR-mutant lung cancer receive chemotherapy, bevacizumab, or other agents alongside their tyrosine kinase inhibitor, and several of those drugs carry their own cardiovascular risks. A retrospective, single-center design cannot disentangle the contribution of each drug, nor can it exclude the possibility that patients who developed toxicity differed systematically in baseline cardiovascular health from those who did not. The case of furmonertinib, where all five assessed patients developed toxicity, is a signal that demands attention precisely because it cannot yet be interpreted as a rate.

What the study does provide is a well-characterized real-world description of the problem, something that regulatory pharmacovigilance databases cannot easily supply. The researchers note that signals of cardiovascular toxicity have been explored using the U.S. Food and Drug Administration’s Adverse Events Reporting System, a database of spontaneous reports that can flag associations but cannot establish incidence or mechanism. A carefully documented cohort with paired echocardiographic and electrocardiographic data, graded by CTCAE criteria, complements those signals with anatomical and physiological specificity. The combination of reduced ejection fraction and QTc prolongation observed here mirrors the dual cardiac liability previously suspected for osimertinib and related agents, and the concurrent decline in chamber dimensions gives mechanistic texture to what has often been reported only as isolated adverse event counts.

For the growing population of patients living with EGFR-mutant lung cancer, the practical message is one of informed vigilance rather than alarm. The absolute incidence of severe cardiovascular toxicity in this cohort was low, at about one percent of treated patients over the study period, and most affected patients recovered with standard heart failure therapy. But because ejection fraction declined measurably even in the cohort as a whole, and because the most serious events clustered among patients on multiple drugs in the class, the study strengthens the argument for baseline echocardiography and electrocardiography before starting third-generation EGFR inhibitor therapy, with serial reassessment during treatment, especially in patients with pre-existing cardiac risk factors. The authors call for larger multicenter prospective studies with standardized monitoring protocols and comparator groups, and until those studies arrive, this single-center snapshot stands as a reminder that a drug designed to be kind to one organ may not always be kind to another, and that the longest cancer journeys deserve cardiac surveillance built into the itinerary from the very first prescription.

Subject of Research: Cardiovascular toxicity during third-generation EGFR tyrosine kinase inhibitor treatment in EGFR-mutant non-small cell lung cancer

Article Title: Cancer therapy-related cardiovascular toxicity observed during third-generation EGFR-TKI-based treatment in EGFR-mutant NSCLC: a retrospective cohort study

Article References: Cancer therapy-related cardiovascular toxicity observed during third-generation EGFR-TKI-based treatment in EGFR-mutant NSCLC: a retrospective cohort study. (n.d.). https://doi.org/10.1186/s12885-026-17095-8

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17095-8

Keywords: EGFR-mutant NSCLC, third-generation EGFR-TKIs, osimertinib, furmonertinib, cardiovascular toxicity, heart failure, QTc prolongation, left ventricular ejection fraction, cardio-oncology, drug safety, retrospective cohort study, CTCAE grading