The study population reflected the realities of cancer presentation in India, and those realities are sobering. The mean age of patients was 70.7 years, with a standard deviation of 5.2 years, and 73 patients, or 21.9 percent, were 75 or older. Men dominated the cohort at 68.8 percent, yielding a male-to-female ratio of 2.2 to 1. Tumors were distributed across the anatomical spectrum of the digestive system: 151 patients, or 45.3 percent, had tumors of the upper gastrointestinal tract, including the esophagus and stomach; 117 patients, or 35.1 percent, had lower gastrointestinal tumors such as colorectal cancers; and 65 patients, or 19.5 percent, had tumors of the hepato-pancreato-biliary system, encompassing the liver, pancreas, and biliary tree. These are among the most lethal and symptomatically silent cancers known to medicine.

What emerges most forcefully from the data is how late these patients arrived at the clinic. At presentation, 242 patients, or 72.7 percent, already had stage III or stage IV disease, meaning their tumors had either spread extensively to regional lymph nodes or metastasized to distant organs. Even more consequential for treatment planning, 206 patients, or 61.9 percent, had a performance status of 3 or higher on the Eastern Cooperative Oncology Group scale, a widely used metric in which scores range from 0, indicating fully functional patients with no restriction of activity, to 5, indicating death. A score of 3 means the patient is capable of only limited self-care and is confined to bed or a chair for more than half of waking hours. A score of 4 means the patient is completely disabled and entirely dependent on others. In practical terms, nearly two-thirds of these patients were too physically depleted to tolerate the rigors of surgery, chemotherapy, or radiation at the moment they first sought care.

Against this backdrop, curative intent was assigned to 169 patients, or 50.8 percent of the cohort, meaning just over half were deemed candidates for treatment aimed at eradicating the disease. The crude, unadjusted statistics initially suggested that age played a role: the curative-intent rate fell from 53.8 percent among patients aged 65 to 74 to 39.7 percent among those aged 75 and older, an odds ratio of 0.56 with a 95 percent confidence interval of 0.33 to 0.96 and a p-value of 0.034. On its face, this looks like evidence of ageism in treatment allocation, or at least of a biological threshold beyond which curative efforts become less likely. But the raw numbers tell only part of the story, and the researchers knew it.

The far more dramatic gradients appeared along clinical rather than chronological lines. Curative intent was assigned to 78.9 percent of patients with a performance status of 0 or 1, the fittest individuals in the cohort, but to a mere 1.1 percent of patients with a performance status of 4. Similarly, curative intent was recorded in 92.3 percent of patients diagnosed at stage I, when the tumor is still localized, but in only 18.9 percent of those diagnosed at stage IV, when distant metastases have taken hold. These gradients are steep, nearly vertical, and they dwarf the modest decline associated with crossing the 75-year threshold. The pattern strongly suggested that older patients were less likely to receive curative treatment not because of their birthdates but because they arrived at the hospital with more advanced disease and more depleted bodies.

To disentangle these intertwined variables, the team employed multivariable binary logistic regression, a statistical technique that estimates the independent effect of each predictor while holding the others constant. The result was decisive. After adjustment for stage and performance status, age of 75 years or older was no longer statistically significant, with an adjusted odds ratio of 0.54, a 95 percent confidence interval of 0.28 to 1.02, and a p-value of 0.059 that straddles the conventional threshold of 0.05. In contrast, stage IV disease carried an adjusted odds ratio of 0.09, and an ECOG performance status of 3 or 4 carried an adjusted odds ratio of 0.16, both with confidence intervals far from unity and p-values below 0.001. In plain language, a patient with metastatic disease was roughly eleven times less likely to receive curative treatment than a comparable patient with earlier-stage disease, and a severely debilitated patient was roughly six times less likely than a fit one, regardless of age.

The statistical robustness of the model lends considerable weight to these conclusions. The area under the receiver operating characteristic curve, a standard measure of a model’s ability to discriminate between outcomes, reached 0.85, which is generally considered excellent, and the McFadden pseudo-R-squared of 0.31 indicates a strong fit for a logistic model of this kind. In other words, knowing a patient’s cancer stage and functional status allows clinicians to predict the treatment decision with a high degree of accuracy, while knowing the patient’s age adds essentially nothing once those two variables are accounted for. The crude age effect observed in the unadjusted analysis was, as the authors put it, a reflection of confounding by stage and performance status: older patients fared worse not because they were old but because their disease was caught later and had taken a greater toll on their bodies.

Perhaps the most alarming finding in the study concerns the gap between recommendation and delivery. Only 117 of the 333 patients, or 35.1 percent, actually received and completed cancer-directed treatment. Even more striking, 129 patients, or 38.7 percent, were advised to undergo treatment but did not accept it. This rate of treatment non-acceptance is a phenomenon with deep roots in the socioeconomic and cultural landscape of Indian healthcare, encompassing financial constraints, distance from treatment centers, fear of surgery and its complications, family decision-making dynamics, and, in some cases, a fatalistic acceptance of the diagnosis. Whatever its causes, the sheer scale of the problem means that nearly four in ten older patients who were offered a chance at cancer-directed therapy walked away from it, a figure that dwarfs any effect attributable to age itself.

The implications of this study ripple outward in several directions. For clinicians, the message is to evaluate older cancer patients on their individual merits, using validated measures of functional status and comorbidity rather than relying on chronological age as a proxy for frailty or tolerance. A vigorous 82-year-old with stage II colon cancer and an ECOG score of 1 may be an excellent surgical candidate, while a depleted 67-year-old with stage IV disease and an ECOG score of 4 may not. For health policymakers, the data point unambiguously toward the need for earlier diagnosis. When 72.7 percent of patients present with stage III or IV disease and 61.9 percent arrive with severely compromised functional status, the battle is largely lost before treatment planning even begins. Screening programs, public awareness campaigns, and improved access to primary care and diagnostic endoscopy could shift the entire distribution of presentations toward earlier, more treatable stages.

For patients and families, the study offers a measure of reassurance grounded in evidence rather than sentiment. Age alone should not be read as a verdict. The researchers argue that their findings argue for earlier diagnosis rather than age-based treatment limitations, and the numbers bear that out: the crude age effect dissolved under statistical adjustment, while stage and performance status remained the true arbiters of curative possibility. As populations in India and across the world age, and as gastrointestinal cancers continue to claim a rising share of the cancer burden among older adults, the challenge for oncology is not to ration care by birth year but to detect disease sooner, to support patients’ functional health, and to dismantle the barriers, financial, geographic, and cultural, that keep nearly four in ten patients from accepting the treatment they are offered. The study, published open access in the Journal of Cancer Research and Clinical Oncology, was approved by an institutional ethics committee with a waiver of individual informed consent for the analysis of de-identified data, and the authors declare no conflicts of interest.

Subject of Research: Determinants of curative versus palliative treatment intent in older adults with gastrointestinal cancer

Article Title: Performance status over chronology: determinants of curative versus palliative treatment intent in older adults with gastrointestinal cancer

Article References: Simon, P., Krishna, A., Banerjee, S., Athiyamaan, M. S., Sunny, J., Srinivas, C., Lobo, D., Suvarna, M., Baruah, P., & Abinesh, E. (2026). Performance status over chronology: determinants of curative versus palliative treatment intent in older adults with gastrointestinal cancer. Journal of Cancer Research and Clinical Oncology. https://doi.org/10.1007/s00432-026-06625-3

Image Credits: AI Generated

DOI: 10.1007/s00432-026-06625-3

Keywords: gastrointestinal cancer, older adults, performance status, ECOG, cancer staging, palliative care, curative intent, geriatric oncology, logistic regression, India, treatment non-acceptance, early diagnosis