Most patients with head and neck cancer see their anxiety, fear of recurrence, and oral pain fade over the years after diagnosis, but a new analysis of more than 5,000 patients reveals a troubling exception: depression tends to worsen as time goes on, climbing steadily across every symptom group the researchers identified. The findings, published in the Journal of Cancer Survivorship, map the long arc of four major symptoms from the moment of diagnosis into long-term survivorship, and they suggest that the psychological weight of this disease may outlast the treatment period that causes so much of the physical burden.
The study drew on the Head and Neck 5000 cohort, one of the largest longitudinal clinical datasets of head and neck cancer patients available for research. Patients completed assessments roughly four months after diagnosis, again at twelve months, and finally between 36 and 60 months, with the average time since diagnosis at the last timepoint standing at 48 months. Initial data collection for the parent cohort took place between 2011 and 2014.
To trace how symptoms evolved, the researchers applied two statistical techniques designed for repeated measures: latent class growth analysis and growth mixture modeling. The models produced strong class separation across all four symptom domains, meaning the identified groups were statistically distinct.
The results showed that the majority of patients fell into mild symptom classes for three of the four domains: 58 percent for depression, 48 percent for anxiety, and 62 percent for oral pain. Fear of cancer recurrence behaved differently, with its largest class, comprising 44 percent of patients, representing moderate symptoms. For anxiety and oral pain, the mild classes showed significantly decreasing slopes over time, marking them as mild-recovery groups. The mild depression class, by contrast, had significantly increasing slopes, making it a mild-worsening group. Across all four symptoms, a minority of between 10 and 18 percent of patients occupied an elevated symptom class, and among elevated depression and anxiety these trajectories were worsening rather than recovering.
Taken across the four symptom domains, nearly 20 percent of patients belonged to an elevated trajectory, a proportion the authors describe as clinically important. These patients were consistently more likely to be younger, to have lower incomes, and to report greater weekly alcohol consumption. For elevated depression, patients with stage III or IV cancer had roughly double the odds of belonging to the elevated class compared with the majority mild class, with an odds ratio of 2.02, while each increment of alcohol consumption raised the odds slightly, at 1.01.
Anxiety and oral pain showed similar patterns with their own nuances. Current or former tobacco users had more than double the odds of belonging to the elevated anxiety class, at an odds ratio of 2.09, and nearly four times the odds for elevated oral pain, at 3.57. Advanced stage disease raised the odds of elevated oral pain even higher, to 3.71, and also elevated the odds for fear of recurrence at 1.62. Female patients, meanwhile, had markedly lower odds of elevated anxiety, at 0.33, though the growth mixture models revealed a counterintuitive detail: within the mild anxiety class, male patients’ anxiety increased faster over time than that of female patients. Alcohol consumption emerged as a universal predictor, significantly raising the odds of elevated class membership for all four symptoms.
HPV status told a more complex story. It did not predict which trajectory class a patient belonged to for any of the four symptoms, but it did shape how symptoms moved within classes. Patients with HPV-positive tumors experienced faster reductions in anxiety and fear of cancer recurrence over time, consistent with the idea that this subgroup may be less psychologically vulnerable at diagnosis, perhaps reflecting their distinct sociodemographic profile. Yet the same patients showed greater increases in depression over time. Within the moderate depression class, those with HPV-positive disease started with lower depression scores at diagnosis but their scores rose at greater rates than for HPV-negative patients. The authors caution against equating improved clinical outcomes with better quality of life, noting that psychosocial support must remain equally accessible for all patients throughout survivorship.
The depression findings carry particular weight because of what is already known about the condition in this population. Depression is highly prevalent among head and neck cancer patients, and unmanaged depression has been linked to decreases in quality of life and to all-cause mortality. The authors suggest the worsening pattern may reflect the unique consequences of the disease, such as functional impairment or social isolation that compound initial distress. They also note that the observed course matches prior head and neck cancer samples, in which anxiety decreases gradually over time while depression, initially lower around diagnosis, may increase during and after treatment completion. Oral pain, meanwhile, remains one of the most prominent and distressing symptoms reported both before diagnosis and after treatment ends, with documented associations to diminished quality of life and increased psychological distress.
The study has limitations the authors are careful to enumerate. The observational design allows inferences about temporal sequence but cannot establish causal mechanisms. The archival cohort offers limited sociodemographic information, few follow-up timepoints, and older data that may not reflect advances in radiotherapy, targeted therapies, treatment de-escalation, and minimally invasive surgery developed since 2014, all of which could reduce long-term symptom burden. The sample was also demographically homogeneous, preventing analysis of race, ethnicity, or sexual and gender identity, even though racial differences in head and neck cancer outcomes have been reported even after accounting for socioeconomic factors. The authors further acknowledge that finite mixture modeling can yield varying results depending on model specification, and that the identified classes should be understood for their qualitative and quantitative utility rather than as true distinct subpopulations. The HADS measure used for anxiety and depression has known difficulties discriminating between symptoms, and the analysis did not cover fatigue, body image disturbance, or functional impairment.
Despite these caveats, the authors argue the findings point toward concrete clinical practice. They advocate longitudinal screening throughout survivorship, using brief validated tools such as the Distress Thermometer or the HADS to identify patients at risk, with screeners guiding referrals to mental health services for those who might not otherwise seek help. They see the results as supporting stepped-care models that match patients to services such as psychological support, physical therapy, or speech pathology according to symptom severity, and they call for high-quality randomized trials of psychosocial interventions tailored to head and neck cancer patients, including approaches like mind-body programs that require less verbal participation, an advantage for patients coping with speech difficulties.
Subject of Research: Long-term trajectories of anxiety, depression, fear of cancer recurrence, and oral pain in head and neck cancer patients
Article Title: Symptom trajectories in head and neck cancer from diagnosis to long-term survivorship
Article References: Fenech, A. L., Park, E. R., Wirth, L. J., D’Souza, G., Humphris, G., Siegel, S. D., Pring, M., & Laurenceau, J.-P. (2026). Symptom trajectories in head and neck cancer from diagnosis to long-term survivorship. Journal of Cancer Survivorship. https://doi.org/10.1007/s11764-026-02146-7
Image Credits: AI Generated
DOI: 10.1007/s11764-026-02146-7
Keywords: head and neck cancer, symptom trajectories, depression, anxiety, fear of cancer recurrence, oral pain, HPV-positive, cancer survivorship, latent class growth analysis, Head and Neck 5000, psychosocial support, alcohol consumption
News Source: Nathaniel Bowman. (November 5, 2026). Head and neck cancer symptoms: most recover, but depression worsens over time. Scienmag.
