By Arunima Rajan

Dr Geoffrey Pelz on how government-funded, risk-based CT screening that extends beyond smokers and reaches people before symptoms appear has moved most screen-detected cancers into the early, treatable stages

Lung cancer in never-smokers is rising worldwide and is now discussed as a distinct disease. How settled is that view, and what is driving it?

Fairly settled. Tumours in never-smokers have their own molecular and clinical signature, often driven by mutations such as EGFR that respond to targeted drugs rather than conventional chemotherapy. The rise has no single cause. Outdoor air pollution, particularly PM2.5, indoor pollutants from cooking fuels and naturally occurring radon are all significant drivers, and genetic predisposition makes some people more susceptible. Secondhand smoke remains a recognised risk. As smoking rates fall, the never-smoker share also becomes more visible. Better detection is not a main driver – if anything, diagnosis in never-smokers is late, because routine screening isn't recommended for them.

Does the evidence show ambient air pollution causes lung cancer, or merely correlates with it? Where are we still extrapolating?

It causes it. The mechanism is well understood: inhaled fine particulates lodge in the lung and trigger chronic inflammation, which wakes dormant cells already carrying cancer-driving mutations. The particles also damage DNA directly and impair repair, producing the genetic instability that fuels tumours. What remains extrapolation is the dose-response curve at the extreme, sustained concentrations found in many Indian cities, and disentangling the local pollutant mixture from co-occurring exposures such as indoor smoke. For readers who live with this permanently rather than episodically, pollution is a confirmed carcinogen. The open question is the size of the risk, not its existence.

Low-dose CT screening rests on the NLST and NELSON trials, both built on heavy-smoker cohorts. What happens to that evidence in a population where many patients never smoked?

It doesn't transfer. Those trials depended on a single powerful variable – pack-years – to decide who to scan. Remove it and the model collapses. A defensible criterion would abandon the smoking-centric approach for a cumulative risk score combining long-term environmental exposure, such as residential history in high-PM2.5 areas and years of biomass fuel use, with older age, family history and prior lung disease including COPD or tuberculosis. Crucially, such a tool cannot simply be imported. It has to be built and validated prospectively in the population it will serve, or the harms of screening may outweigh the benefits.

What is the realistic minimum a health system needs for molecular testing to be meaningful? At what point does testing become a false promise?

Three things, inseparable: a focused panel, fast results, and the drugs to act on them. The minimum is not an expensive comprehensive panel but a targeted one covering the common actionable biomarkers – EGFR, ALK, ROS1 and PD-L1 – reported within two weeks, so first-line treatment isn't delayed. The critical element is the third. A positive result must lead to affordable access to the corresponding drug. Testing becomes a false promise the moment that link breaks: you hand a patient a diagnosis you cannot act on, spend scarce resources on unusable information, and widen existing inequities in care.

Which countries have genuinely shifted stage at diagnosis, and how? I want mechanisms, not aspirations.

Taiwan and the UK. The UK's targeted lung health check, now national, invites a broad population for risk assessment and offers low-dose CT to those found to be high-risk; more than 76% of the cancers it finds are early stage. Taiwan's national programme is more adaptable to diverse populations because it deliberately includes high-risk never-smokers on the basis of documented family history – a data-driven choice, and 85% of screen-detected cancers there are stage 0 or I. The transferable mechanisms: government-funded, risk-stratified CT screening that looks beyond smoking history; local data to define non-smoking risk groups; and proactive, population-based recruitment before symptoms appear.

Which access mechanisms for targeted therapy and immunotherapy have actually worked in middle-income countries, and which have been public relations?

What works changes market structure. Biosimilars and generics create durable competition and permanently lower prices for the whole system. Compulsory licensing, or the credible threat of it, breaks monopolies and forces steep reductions. What mostly doesn't: patient assistance programmes and opaque tiered pricing. Assistance schemes can be a lifeline for an individual, but they are not systemic, and they often function as a way of building demand for a high-priced drug. Tiered pricing routinely offers discounts too small to make a drug affordable to a public system, while preserving the manufacturer's pricing power and reputation.

What did high-income systems get wrong in building lung cancer services that a country still building them could skip?

They built in sequential, treatment-focused silos. A newer system can skip that by starting diagnosis-led: a mandatory multidisciplinary team as the coordinating hub from the first appointment, and a strict rule that molecular results precede treatment, rather than defaulting to chemotherapy and correcting later. It can also skip the era of repeat biopsies by designing a pathway that takes enough tissue for every test in a single procedure. And it can avoid treating palliative care as an afterthought, integrating it from diagnosis alongside active treatment rather than when treatment runs out.

One investment in a resource-constrained system: imaging, pathology, a tumour board, palliative care? Honest answer?

A functioning, mandatory multidisciplinary tumour board. It is the cheapest high-leverage intervention available and there is solid evidence it improves survival. Unlike a single piece of equipment, it works as an operating system: it forces specialists to communicate, collapses the delays that kill people, corrects diagnostic and staging errors in up to 40% of cases, and raises adherence to evidence-based care. Imaging, pathology and palliative care all matter enormously, but their value is only realised when something coordinates them – ensuring the right tests are ordered, the results acted on, and supportive care offered from the start.

Got a story that Healthcare Executive should dig into? Shoot it over to arunima.rajan@hosmac.com—no PR fluff, just solid leads.

How Taiwan and the UK are finding lung cancer earlier