Anxiety and Depression May Blunt Immunotherapy Response in Stomach Cancer Patients

Emotional distress before surgery may quietly undermine one of the most promising treatments for stomach cancer. A new retrospective cohort study from two Chinese hospitals suggests that patients with locally advanced gastric cancer who show symptoms of depression or anxiety respond far less well to neoadjuvant immunotherapy than their psychologically healthier counterparts. The findings, published in BMC Cancer, add to a growing body of evidence that the mind and the immune system are deeply intertwined, and that a patient’s emotional state could influence whether cutting-edge cancer drugs actually work.

The research team, led by Cheng Wu, Xunjun Li, Jiawen Chen, Chuanfa Fang and Tao Chen of Southern Medical University’s Nanfang Hospital system, examined 129 patients with locally advanced gastric cancer who received anti-PD-1 based immunotherapy before surgery between April 2022 and December 2024. Neoadjuvant therapy, given before the primary surgical treatment, has become a cornerstone approach for shrinking tumors and improving surgical outcomes in this aggressive disease. Immune checkpoint inhibitors such as anti-PD-1 antibodies work by releasing the molecular brakes that tumors place on T cells, allowing the body’s own immune defenses to attack cancer cells. But these drugs do not work for everyone, and oncologists have been searching for the factors that separate responders from non-responders.

To measure emotional distress, the researchers used two widely validated screening instruments: the Patient Health Questionnaire-9, or PHQ-9, which assesses depressive symptoms, and the Generalized Anxiety Disorder-7 scale, known as GAD-7. Patients scoring five or higher on either scale were classified as experiencing emotional distress. By this definition, 38 percent of the cohort, 49 of 129 patients, met the threshold for distress. That prevalence is consistent with what oncologists routinely observe in cancer wards, where a diagnosis of gastric cancer, often discovered at an advanced stage in populations with high disease burden, takes a substantial psychological toll.

The difference in treatment response between the two groups was striking. Among patients with emotional distress, only 36.7 percent showed a meaningful pathological response to the immunotherapy, as graded by the College of American Pathologists Tumor Regression Grade system, which pathologists use to measure how much tumor tissue has been destroyed by preoperative treatment. Among patients without distress, the response rate was 81.3 percent. The gap was statistically robust, with a P value below 0.001, meaning the probability that such a difference arose by chance alone is extremely small.

Because retrospective comparisons can be confounded by other variables, the team went further, applying multivariate logistic regression to adjust for factors that might independently influence response. Even after this statistical correction, emotional distress remained an independent negative predictor of pathological response, with an adjusted odds ratio of 0.252 and a 95 percent confidence interval of 0.076 to 0.809, and a P value of 0.021. In practical terms, distressed patients had roughly a quarter of the odds of responding to therapy compared with non-distressed patients. The authors caution, as they should, that retrospective designs cannot prove causation, but the association survived the standard statistical hurdles that such studies must clear.

Perhaps the most intriguing clue in the study concerns perineural invasion, or PNI, a pathological feature in which cancer cells invade and travel along nerves. PNI is recognized as an aggressive behavior in many cancers and is associated with worse prognosis. In this cohort, emotional distress correlated strongly with higher PNI prevalence: 77.8 percent of distressed patients showed perineural invasion, compared with only 26.5 percent of non-distressed patients, again with a P value below 0.001. This correlation offers a possible biological bridge between mood and treatment outcome, suggesting that distress may be linked to tumor biology that is inherently more invasive and less responsive to immune attack.

The mechanistic story that scientists have been assembling over decades makes such a link biologically plausible. Chronic psychological stress activates two major physiological systems: the hypothalamic-pituitary-adrenal axis, which drives sustained cortisol secretion, and the sympathetic nervous system, which floods tissues with catecholamines such as norepinephrine. Both pathways can suppress immune function, dampening T cell activity, reshaping the tumor microenvironment, and promoting inflammation patterns that favor cancer progression. Nerve-cancer interactions, the phenomenon underlying perineural invasion, may themselves be amplified by stress-related neural signaling, since tumors and nerves engage in a chemical dialogue that can stimulate tumor growth and dissemination. If emotional distress tilts this dialogue toward the tumor, it could simultaneously worsen invasiveness and blunt the effectiveness of drugs that depend on a vigorous immune response.

An important subgroup analysis strengthens the clinical relevance of the findings. PD-L1, a protein that tumors use to suppress immune cells, is measured by the combined positive score, or CPS, and a CPS of five or higher is a well-established biomarker predicting benefit from anti-PD-1 therapy. Patients whose tumors express high PD-L1 are expected to be the best responders to checkpoint inhibitors. Yet even within this favorable PD-L1 CPS of five or greater subgroup, patients without emotional distress maintained superior treatment response compared with distressed patients. In other words, a biomarker that normally identifies patients likely to benefit from immunotherapy did not fully protect distressed individuals from the negative association, hinting that psychological state may carry prognostic information that standard biomarkers miss.

The implications for clinical practice are significant. If the association holds up in prospective research, screening gastric cancer patients for depression and anxiety before they begin neoadjuvant immunotherapy could become as routine as measuring PD-L1 expression. Psychological support, whether through counseling, cognitive behavioral therapy, stress reduction programs or pharmacological treatment, might then be positioned not merely as an adjunct to comfort care but as a component of cancer treatment with measurable biological stakes. The study authors explicitly highlight the need for psychological support in this population and call for prospective studies to validate the underlying mechanisms and determine whether treating distress can actually improve immunotherapy outcomes.

As with all retrospective research, important caveats apply. The study included 129 patients from two Chinese hospitals, and the cohort’s characteristics may not generalize to all populations. Emotional distress was measured with screening questionnaires rather than formal psychiatric diagnosis, and the timing of the assessments relative to treatment could influence the results. Reverse causation remains possible: patients whose tumors are biologically more aggressive might feel worse, rather than the other way around, although the correlation with perineural invasion suggests a genuine biological entanglement. Still, the consistency of the findings, the strength of the statistical associations, and the plausible mechanistic framework make this one of the more compelling studies to date linking emotional state to immunotherapy efficacy in solid tumors. For the roughly one in three gastric cancer patients experiencing distress, the message is a call for integrated care, where oncologists, surgeons and mental health professionals work together, because the battle against cancer may be fought not only with antibodies and scalpels but also with the full cooperation of a patient’s mind and immune system.

Subject of Research: Association between emotional distress and pathological response to neoadjuvant immunotherapy in locally advanced gastric cancer

Article Title: Emotional distress and pathological response to neoadjuvant immunotherapy in locally advanced gastric cancer: a multicentre retrospective cohort study

Article References: Wu, C., Li, X., Chen, J., Fang, C., & Chen, T. (2026). Emotional distress and pathological response to neoadjuvant immunotherapy in locally advanced gastric cancer: a multicentre retrospective cohort study. BMC Cancer. https://doi.org/10.1186/s12885-026-17035-6

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17035-6

Keywords: gastric cancer, neoadjuvant immunotherapy, emotional distress, anti-PD-1, perineural invasion, pathological response, immune checkpoint inhibitors, PHQ-9, GAD-7, PD-L1 CPS, psychological stress, tumor microenvironment