I see a particular kind of patient often enough now that the pattern has stopped surprising me. A woman in her early forties, schoolteacher or homemaker or junior manager, never smoked, never lived with anyone who smoked, comes to clinic with a cough that has lasted three months. She has been to two general physicians and one chest specialist before me. She has been treated for tuberculosis empirically. She has been told it is allergy. She has been told it is post-Covid sequelae. By the time she walks in, the imaging done a week earlier shows a mass.
The diagnosis is lung cancer. The reason it took three months to get there is that nobody, including her, believed lung cancer was on the list of things this could be. She did not smoke. The reasoning ended there.
The pattern is well-documented
Across India and East Asia, the demographic profile of lung cancer is shifting. The classic patient — a male smoker in his sixties — is still the most common globally. But in Indian metros, especially among women under fifty, the dominant lung cancer presentation is now in non-smokers. Some recent series put the figure above fifty per cent. Several factors are driving this. Ambient air pollution in Indian cities, especially winter PM 2.5 levels, is now classified as a Group 1 carcinogen by the IARC. Indoor smoke from biomass stoves and kerosene burners — still common in semi-urban and rural homes — adds to the load. Occupational exposures in construction, mining, welding, and the chemical industry are under-appreciated. And there is a meaningful genetic component: EGFR mutations, which drive a substantial share of non-smoker lung adenocarcinomas, are far more common in Asian populations than in Western ones.
None of this is new science. What is new is the speed at which the demographic has shifted. The mental model in most general clinics still pattern-matches "lung cancer = smoker." That model is now wrong often enough that it costs people months of diagnostic delay.
The cost of delay is measurable
Lung cancer is a stage-sensitive disease. Caught at stage I, five-year survival approaches sixty to seventy per cent. Caught at stage IV, five-year survival is in single digits for most subtypes. The difference between stage I and stage III is often a matter of months — and the months that matter are usually the months before diagnosis, not after.
The most common cause of avoidable diagnostic delay in non-smokers is exactly what played out for the patient I described above. Persistent cough is treated empirically with antibiotics, then with anti-tuberculosis therapy, then with inhalers, while the underlying mass continues to grow. By the time imaging is finally ordered, the cancer is no longer early-stage.
This is not a story about a missed diagnosis from negligence. It is a story about a heuristic that needs updating. The heuristic to update is simple: any cough that lasts more than three weeks should be evaluated, regardless of smoking status. The investigation does not need to be elaborate. A chest X-ray and clinical examination are enough to begin. Most prolonged coughs are not cancer. The point is that the small fraction that are need to be caught early.
The patient with a persistent cough who 'doesn't smoke' is the patient most often delayed to diagnosis.
What to actually look for
Beyond persistent cough, the warning signs of lung cancer haven't changed. They're worth restating because non-smokers and their families often dismiss them — wrongly assuming the symptoms must mean something else.
- Cough lasting more than three weeks that does not respond to standard treatment, regardless of smoking history.
- Any episode of coughing up blood — even small amounts, even once. Hemoptysis is always a reason to seek evaluation.
- Unexplained weight loss — more than five kilograms over a few months without a change in diet or activity.
- Persistent breathlessness, particularly with exertion that previously didn't trouble the person.
- Recurrent chest infections — pneumonia in the same lung area twice in a year warrants a CT scan, not just another antibiotic course.
- Persistent chest pain, hoarseness of voice, or new clubbing of the fingers.
Why diagnosis matters more than ever
The reason urgency around early diagnosis has grown over the last decade is that lung cancer treatment has become dramatically more effective — but only for patients diagnosed at stages where treatment is still curative-intent or controllable.
For patients with EGFR-mutated, ALK-rearranged, ROS1, KRAS G12C, or other targetable subtypes, oral targeted therapies often outperform chemotherapy as first-line treatment. Modern agents like osimertinib (for EGFR), alectinib (for ALK), and sotorasib (for KRAS G12C) routinely produce responses lasting two to three years or more in advanced disease — and the field is moving these drugs into the adjuvant setting after surgery, where they appear to extend disease-free survival meaningfully.
For patients without targetable mutations, immunotherapy — pembrolizumab, atezolizumab, durvalumab, nivolumab — has changed the playing field, especially in PD-L1-high disease. The PACIFIC regimen (concurrent chemoradiation followed by durvalumab consolidation) has become the standard of care for unresectable stage III non-small-cell lung cancer.
None of these gains help a patient who is diagnosed at stage IV with a brain metastasis, three months after her cough began. Many of them help — sometimes spectacularly — a patient diagnosed at stage I or II.
Screening: where we are, where we're going
Low-dose CT lung cancer screening is recommended internationally for heavy smokers aged 50 to 80, typically those with a 20+ pack-year history. In high-resource settings, this programme has reduced lung cancer mortality by around twenty per cent in screened populations.
For non-smokers, evidence-based screening guidelines are still emerging. There is no general-population screening test for non-smoker lung cancer in India. Several institutions, including ours, are watching the data on whether risk-stratified screening — based on family history, occupational exposure, EGFR family-cluster patterns, or CT abnormalities found incidentally — should be offered to selected non-smokers in high-pollution metros. For now, what every adult should know is the symptom triad of persistent cough, hemoptysis, and unexplained weight loss, and what to do about it.
If you live in a city like Hyderabad and you've had a cough for more than three weeks, it is worth a chest X-ray and a doctor's examination. Smoker or not.
The bottom line
Lung cancer in India is no longer primarily a smoker's disease. The mental model needs to update with the demographics. For patients, the practical implication is straightforward: a persistent cough deserves evaluation, not a third antibiotic course. For doctors and family physicians, the practical implication is the same in mirror: the patient who doesn't smoke and has a three-week cough is exactly the patient who is most often diagnosed late. Catching her at stage I instead of stage III is a months-of-attention problem, not a years-of-research problem. The science has done its part. What's left is the recognition.