Precision oncology has transformed modern cancer care by shifting treatment from organ-based classifications to molecular and genomic insights, promising the right treatment for the right patient at the right time. 

Advances in multi-omics, computational biology and biomarker-driven stratification have reshaped how cancers are diagnosed and treated.

Precision medicine promises individualised care, but only if the models behind it truly represent the patients it serves.

Yet, beneath this progress lies a critical paradox. Despite its sophistication, much of precision medicine is still built on non-representative population foundations. 

Most preclinical cancer models that inform drug discovery are derived from a narrow subset of populations, with Asian and particularly Southeast Asian patients critically underrepresented.

This poses a deeper challenge for the field: How can we truly advance the vision of “United by Unique” – the theme for this year’s World Cancer Day – when the molecular signatures of entire populations are not represented in the models that shape precision care? 

Until population diversity is embedded at the earliest stages of research, precision medicine risks being selectively precise rather than universally transformative.

Precision medicine cannot be truly precise if entire populations are missing at the foundation of discovery.

Preclinical models: Where precision begins or breaks

Preclinical models are the invisible architects of clinical outcomes. From two-dimensional cell lines to three-dimensional spheroids, organoids and patient-derived models, they shape how we understand tumour biology and treatment response, including most anticancer drugs.

What happens in the lab determines what reaches the clinic.

When these models fail to reflect real-world genetic and biological diversity, translational accuracy suffers. Drugs may appear effective in the laboratory yet fail in patients, not only because cancer is complex, but because the models themselves were never representative. 

The critical question, therefore, is not whether precision oncology works, but for whom it works best.

The “one-size-fits-all” fallacy in precision medicine

Although precision medicine is meant to move beyond “one-size-fits-all” care, many molecular signatures and treatment strategies are still applied broadly across populations. 

This limitation is especially clear in highly heterogeneous cancers such as brain cancer, where even small population-level differences can profoundly influence disease behaviour and treatment response, challenging the idea that a single precision framework fits all patients.

In our recent work, we identified a 10-gene signature that stratifies glioblastoma patients into high and low-risk groups and consistently predicts survival outcomes across large international datasets. 

While robust, these models were validated predominantly in Western and Chinese cohorts. Their relevance to Southeast Asian populations and to Malaysia specifically remains unknown.

A model that works well for some populations may still fail others if population context is ignored.

Without population-aware validation, even robust molecular models risk being selectively precise – highly effective for some populations, but less accurate for others.

Population precision as the next frontier

Scientific progress often accelerates when models are reimagined. In breast cancer research, three-dimensional spheroids and organoids transformed how tumour architecture and drug response are studied. In triple-negative breast cancer (TNBC), refined preclinical platforms have revealed new molecular drivers and therapeutic vulnerabilities that were previously invisible.

Better models don’t just refine answers; they uncover new biology.

Our group’s work in TNBC spheroid modelling illustrates this principle. By overcoming longstanding challenges in spheroid formation and stability, we enabled more biologically-relevant models for studying tumour behaviour and drug response, opening pathways towards more advanced TNBC patient-based organoid development.

Brain cancer research now stands at a similar inflection point.

From local models to global impact

Our ongoing work, PRIME-GBM, focuses on developing Malaysian patient-derived brain cancer models using locally-sourced clinical specimens. 

Through collaboration with neurosurgical teams at University Malaya Medical Centre and colleagues at Monash University, and by integrating genomics and proteomics, these models aim to capture population-specific disease biology in an underrepresented Southeast Asian population, strengthening translational fidelity and enabling more precise oncology strategies.

Local models are not about geography. They’re about biological accuracy.

Malaysia’s unique multiracial population offers a powerful scientific advantage to disentangle molecular and contextual drivers of cancer biology within a single national cohort – an opportunity rarely available within one shared healthcare and environmental system.

Our diversity is not a limitation. It’s a unique strength that allows us to test precision medicine more honestly.

Why this matters for the future of precision medicine

Precision oncology has reached a crossroads. While technology continues to advance, the field must confront structural biases embedded within its foundational models.

Population-representative models do not weaken precision medicine. They strengthen it.

To truly fulfil the spirit of #CloseTheCareGaps and advance #UnitedByUnique, precision oncology must reflect the biological diversity of the people it aims to serve, to better withstand real-world clinical complexity.

Precision medicine without population precision is incomplete.

By embracing our unique genetic and societal diversity, we can be truly united to build a more accurate and inclusive future for cancer care, one where innovation doesn’t stop at discovery, but reaches every patient who needs it.

When we embrace population precision, we move closer to a future where precision cancer care is not just advanced, but inclusive and transformative.