Cancer Quietly Dampens the Heart’s Hidden Rhythm, Major Analysis Finds

The human heart never beats with metronomic precision. Between every two beats, tiny fluctuations in timing—measured in milliseconds—reveal a constant negotiation between the sympathetic nervous system, which accelerates the heart under stress, and the parasympathetic vagus nerve, which slows it during rest. Clinicians call this fluctuation heart rate variability, or HRV, and for decades it has served as a non-invasive window into the autonomic nervous system. Now, a systematic review and meta-analysis published in Supportive Care in Cancer has pooled evidence from dozens of studies and reached a striking conclusion: cancer patients, across multiple tumor types, show consistently suppressed short-term HRV compared with people without cancer. The finding, while robust in direction, comes with important scientific caveats that the authors themselves are careful to emphasize.

The research team, led by Hsiu-Hua Shih and Tsai-Wei Huang of Taipei Medical University, searched five major biomedical databases—PubMed, MEDLINE, CINAHL, Embase, and the Cochrane Library—through March 2026. Their sweep identified 28 studies encompassing 3,232 participants and spanning nine categories of cancer. To be eligible, studies had to include adults with physician-confirmed malignancy and HRV measured either by electrocardiography or by photoplethysmography, the optical sensing technology embedded in most modern smartwatches and fitness bands. The breadth of the evidence base matters, because HRV research in oncology has long been fragmented across individual small trials, leaving clinicians without a clear quantitative picture of how large the autonomic differences between cancer patients and healthy controls actually are.

A key methodological decision shaped the entire analysis. Time-domain HRV indices such as SDNN, the standard deviation of normal-to-normal heartbeat intervals, scale directly with the length of the recording: a 24-hour ambulatory measurement yields systematically larger values than a five-minute resting snapshot. Pooling studies with different recording durations would therefore conflate measurement artifacts with genuine biology. The researchers restricted their primary quantitative pooling—designated Part A of the review—to short-term resting recordings of five to ten minutes obtained with age-matched non-cancer controls. Only five studies met these strict criteria. Long-term ambulatory recordings of twelve to thirty hours were reported separately, and the remaining studies were synthesized narratively or examined for longitudinal change over time. The team used the DerSimonian–Laird random-effects model and graded the certainty of all findings with the GRADE framework, registering the protocol in advance as PROSPERO CRD42026130103.

The pooled results were unambiguous in direction. Cancer patients showed lower SDNN than controls by a mean difference of 14.05 milliseconds, with a 95 percent confidence interval spanning 9.90 to 18.20 milliseconds. RMSSD, the root mean square of successive heartbeat-interval differences and a sensitive marker of vagal parasympathetic activity, was lower by 16.76 milliseconds. pNN50, the percentage of consecutive heartbeat intervals differing by more than 50 milliseconds, was reduced by 6.05 percentage points. All three differences were statistically significant at p less than 0.001, and heterogeneity across studies was modest, with I-squared values between 22.1 and 33.7 percent—remarkably low for a meta-analysis of clinical studies. Leave-one-out sensitivity analyses, in which each study is removed in turn to test whether any single dataset drives the result, confirmed the robustness of the pooled estimates.

Not every HRV metric told the same story. High-frequency power, a frequency-domain index that also reflects parasympathetic modulation, showed no statistically significant difference between cancer patients and controls, with a p-value of 0.118. The divergence between suppressed time-domain indices and an unchanged high-frequency component is scientifically intriguing, and it underscores a broader point the authors stress: HRV is not a single number but a family of indices, each capturing different aspects of autonomic regulation and each vulnerable to different technical and physiological confounders. Time-domain measures computed over short recordings are particularly sensitive to overall beat-to-beat variability, while spectral measures depend heavily on breathing patterns, recording conditions, and the mathematical assumptions of the underlying algorithms.

Beyond the between-group comparison, the review examined how HRV relates to one of the most common and debilitating symptoms in oncology: cancer-related fatigue. Within individual cancer cohorts, studies consistently reported inverse correlations between HRV and fatigue severity—lower variability, greater exhaustion. This pattern aligns with a physiological substrate proposed by earlier work, including studies of breast cancer survivors showing that low HRV tracks with persistent fatigue long after treatment ends. If reduced vagal tone is genuinely linked to the fatigue experienced by millions of cancer survivors, HRV could eventually help clinicians identify patients at risk and monitor responses to interventions such as exercise programs, which prior meta-analytic work has suggested may partially reverse autonomic dysfunction in this population.

The longitudinal strand of the review adds a dynamic dimension. Studies that followed patients over time documented HRV trajectories across the cancer journey, from preoperative measurements through surgery, chemotherapy, and radiation, into survivorship. Perioperative vagal withdrawal, the cardiotoxic effects of certain chemotherapeutic agents, and the physiological burden of the malignancy itself can each leave distinct imprints on autonomic regulation. Some studies even explored whether HRV patterns differ by tumor location or disease progression, and whether nonlinear HRV measures—fractal and entropy-based indices known to predict cardiovascular mortality in other populations—might carry prognostic information in patients with brain metastases, gastric cancer, or colorectal cancer. These exploratory findings remain preliminary, but they sketch a research agenda in which autonomic monitoring could complement existing clinical assessment.

Yet the authors are emphatic about what the pooled estimate cannot do. Because the aggregate combines mechanistically distinct sources of HRV suppression—the malignancy itself, treatment-related cardiotoxicity, perioperative vagal withdrawal, and long-term survivorship remodeling—the pooled figure cannot attribute the observed suppression to any single cause. Moreover, GRADE assessment rated all pooled outcomes as Very Low certainty, the weakest tier on the five-level scale, reflecting limitations in study design, risk of bias, and imprecision. On this basis, the team explicitly declines to recommend HRV as an established clinical biomarker for cancer care. The distinction between a consistent research signal and a validated clinical tool is one that the field has sometimes blurred, and this review draws the line carefully.

The path forward, according to the authors, requires studies stratified by treatment phase, standardized measurement protocols, and the incorporation of nonlinear HRV indices alongside conventional time-domain and frequency-domain metrics. Wearable devices, which have already been used to classify cancer-related fatigue from HRV signals in recent observational studies, could make large-scale, longitudinal autonomic monitoring feasible outside the clinic. If future research confirms that HRV suppression tracks with symptom burden, treatment toxicity, or survival, the humble millisecond-to-millisecond wobble of the heartbeat could evolve from a research curiosity into a practical, non-invasive vital sign for oncology. For now, the message of this meta-analysis is one of cautious confirmation: the autonomic nervous system is measurably altered in cancer, the alteration is real and reproducible in short-term recordings, and understanding what it means for patients is the next great challenge.

Subject of Research: Heart rate variability suppression in cancer patients versus non-cancer controls: a systematic review and meta-analysis

Article Title: Heart rate variability suppression in cancer patients versus non-cancer controls: a systematic review and meta-analysis

Article References: Shih, H.-H., Chang, H.-C., Chen, P.-H., Chang, C.-L., Chuang, Y.-H., Lai, J.-H., Gautama, M. S. N., & Huang, T.-W. (2026). Heart rate variability suppression in cancer patients versus non-cancer controls: a systematic review and meta-analysis. Supportive Care in Cancer, 34(10), Article 1034. https://doi.org/10.1007/s00520-026-11229-1

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11229-1

Keywords: Heart, rate, variability, suppression, cancer, patients, versus, non-cancer, controls, systematic, review, meta-analysis