Hidden Comorbidity Patterns Shape Cancer Treatment Decisions, Malaysian Study Reveals

When oncologists decide whether a patient with breast or colorectal cancer can safely receive chemotherapy, surgery, or targeted therapy, they have long relied on a deceptively simple tool: a count of coexisting conditions, often condensed into an index such as the Charlson Comorbidity Index. A new study from Malaysia challenges that convention, showing that the specific pattern of comorbidities—not merely their number—carries decisive information about how cancer patients function and how they are treated. The findings, published in Supportive Care in Cancer, suggest that two patients with identical comorbidity scores may face radically different treatment trajectories depending on which conditions cluster together in their bodies.

The research team, led by Yee-Yin Hoo and Nirmala Bhoo-Pathy of Universiti Malaya, together with collaborators from Flinders University, UNSW Sydney, and hospitals across Malaysia, applied a statistical technique called latent class analysis (LCA) to data from 1,117 adults with breast or colorectal cancer drawn from two Malaysian cohorts. Unlike traditional approaches that treat every combination of conditions as equivalent, LCA is a person-centered method that identifies hidden subgroups—latent classes—within a population based on the co-occurrence of characteristics. In this case, the researchers grouped each patient’s record-verified comorbidities into 20 categories and let the algorithm reveal whether distinct patterns of illness existed beneath the surface of the raw counts.

The results were striking. Overall, 695 patients—62.2 percent of the cohort—had at least one comorbidity, with hypertension, dyslipidaemia, diabetes, and obesity emerging as the most prevalent conditions. But the latent class analysis uncovered something far more informative than prevalence figures: three distinct comorbidity profiles. The largest group, comprising 723 patients, had low comorbidity. A second group of 368 patients displayed a clustered cardiovascular-kidney-metabolic (CKM) pattern, combining heart, kidney, and metabolic conditions. A third, much smaller group of just 26 patients exhibited a multisystem pattern, with conditions spread across multiple organ systems.

What makes these findings particularly compelling is how differently the three groups fared despite superficially similar index scores. All 26 patients in the multisystem group and half of those in the CKM group had a modified Charlson Comorbidity Index of 3 or higher—scores conventionally interpreted as indicating substantial comorbidity burden. Yet the two groups behaved very differently in clinical terms. The CKM group was considerably older than the low-comorbidity group, with a mean age of 62 years compared to 51 years, but their ECOG performance status and their receipt of cancer treatment were statistically indistinguishable from the low-comorbidity class.

The multisystem group told a different story entirely. These patients had nearly four times the odds of having poor performance status, defined as an ECOG score of 2 or higher, compared to the low-comorbidity group, with an adjusted odds ratio of 4.04 (95 percent confidence interval 1.50–10.84). More dramatically, they had only one-tenth the odds of receiving chemotherapy (adjusted odds ratio 0.10, 95 percent confidence interval 0.04–0.25) and roughly one-seventh the odds of receiving anti-HER2 therapy (adjusted odds ratio 0.14, 95 percent confidence interval 0.02–0.86). These are not marginal differences; they represent a fundamental divergence in how the healthcare system responds to patients whose illnesses are distributed across multiple organ systems rather than clustered within one metabolic pathway.

The implications reach beyond Malaysia. Comorbidity is conventionally summarized as a count or index—a practice that, as the authors note, treats different combinations of conditions as though they were interchangeable. Prior systematic reviews have found no gold standard for measuring comorbidity in cancer populations, and the Charlson Comorbidity Index, while widely used, has been criticized for its clinimetric limitations. By demonstrating that pattern matters, the Malaysian study adds a new dimension to an ongoing international debate about how to capture the true complexity of patients with cancer who also carry other chronic diseases.

Particularly noteworthy is the finding that patients in the CKM class—despite carrying cardiovascular and renal comorbidities that would traditionally raise red flags—had performance status and treatment receipt similar to those with low comorbidity. This is a double-edged result. On one hand, it suggests that clinicians in this setting are not withholding standard cancer therapies from patients with heart, kidney, and metabolic conditions, which is reassuring from an equity perspective. On the other hand, as the authors emphasize, it highlights risks that are not reflected by performance status alone. Cardiovascular and renal comorbidities can influence drug dosing, increase the likelihood of treatment-related toxicity, and elevate long-term cardiovascular mortality, even when a patient appears functionally fit at diagnosis.

The concept of cardiovascular-kidney-metabolic health has gained substantial momentum in recent years, notably through a presidential advisory from the American Heart Association that framed CKM syndrome as an integrated continuum rather than a collection of separate diseases. Recent real-world data analyses have shown that CKM syndrome stage predicts cardiovascular outcomes in cancer patients, and international cardio-oncology guidelines from the European Society of Cardiology now explicitly address the management of patients with cancer who have pre-existing cardiovascular conditions. The Malaysian study extends this framework into a multiethnic Asian setting, where the epidemiology of cardiometabolic disease differs markedly from that of high-income Western populations.

Malaysia itself provides a compelling backdrop for this research. As an upper-middle-income country with a multiethnic population comprising Malay, Chinese, and Indian communities, Malaysia occupies an important position in global cancer epidemiology. Southeast Asia has seen rapidly rising cancer incidence, and the region’s health systems face the dual burden of infectious and chronic non-communicable diseases. The two parent cohorts—the Happiness Registry, funded by Novartis Malaysia, focusing on advanced breast cancer, and a colorectal cancer cohort supported by Malaysia’s Long-Term Research Grant Scheme—were designed to capture real-world patterns of care in this setting. The funders had no role in the study design, analysis, or reporting of the present secondary analysis.

The authors are careful to note limitations. The multisystem class, with only 26 patients, requires validation in independent cohorts before its associations can be considered definitive. The study also relied on record-verified comorbidities grouped into 20 categories, which, while more granular than many prior studies, still involves clinical judgment about how conditions should be classified. Nevertheless, the core message stands: comorbidity patterns revealed differences that counts alone would have missed. For a field that has long relied on blunt indices to guide treatment decisions, the study offers a methodological upgrade—one that could ultimately help clinicians identify which patients with cancer and comorbidity need closer monitoring, which need modified treatment plans, and which can proceed with standard care despite carrying substantial additional disease burden.

Subject of Research: Comorbidity patterns and their associations with performance status and treatment receipt in breast and colorectal cancers in a multiethnic Asian setting: a latent class analysis

Article Title: Comorbidity patterns and their associations with performance status and treatment receipt in breast and colorectal cancers in a multiethnic Asian setting: a latent class analysis

Article References: Hoo, Y.-Y., Ng, H. S., Danaee, M., Koczwara, B., Ramalingam, S., Subramaniam, S., Abdul Satar, N. F., Thiagarajan, M., Yip, C. H., & Bhoo-Pathy, N. (2026). Comorbidity patterns and their associations with performance status and treatment receipt in breast and colorectal cancers in a multiethnic Asian setting: a latent class analysis. Supportive Care in Cancer, 34(11), Article 1106. https://doi.org/10.1007/s00520-026-11338-x

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11338-x

Keywords: Comorbidity, patterns, associations, performance, status, treatment, receipt, breast, colorectal, cancers, multiethnic, Asian