Contribute Try STAT+ Today

Sharon Begley died of complications of lung cancer on Jan. 16, just five days after completing this article. She was a never-smoker.

Breast cancer wouldn’t have surprised her; being among the 1 in 8 women who develop it over their lifetime isn’t statistically improbable. Neither would have colorectal cancer; knowing the risk, Mandi Pike “definitely” planned to have colonoscopies as she grew older.

But when a PET scan in November 2019 revealed that Pike, a 33-year-old oil trader, wife, and mother of two in Edmond, Okla., had lung cancer — she had been coughing and was initially misdiagnosed with pneumonia — her first reaction was, “but I never smoked,” she said. “It all seemed so surreal.”


Join the club. Cigarette smoking is still the single greatest cause of lung cancer, which is why screening recommendations apply only to current and former smokers and why 84% of U.S. women and 90% of U.S. men with a new diagnosis of lung cancer have ever smoked, according to a study published in December in JAMA Oncology. Still, 12% of U.S. lung cancer patients are never-smokers.

Scientists disagree on whether the absolute number of such patients is increasing, but the proportion who are never-smokers clearly is. Doctors and public health experts have been slow to recognize this trend, however, and now there is growing pressure to understand how never-smokers’ disease differs from that of smokers, and to review whether screening guidelines need revision.

“Since the early 2000s, we have seen what I think is truly an epidemiological shift in lung cancer,” said surgeon Andrew Kaufman of Mount Sinai Hospital in New York, whose program for never-smokers has treated some 3,800 patients in 10 years. “If lung cancer in never-smokers were a separate entity, it would be in the top 10 cancers in the U.S.” for both incidence and mortality.


A 2017 study of 12,103 lung cancer patients in three representative U.S. hospitals found that never-smokers were 8% of the total from 1990 to 1995 but 14.9% from 2011 to 2013. The authors ruled out statistical anomalies and concluded that “the actual incidence of lung cancer in never smokers is increasing.” Another study that same year, of 2,170 patients in the U.K., found an even larger increase: The proportion of lung cancer patients who were never-smokers rose from 13% in 2008 to 28% in 2014.

“It is well-documented that approximately 20% of lung cancer cases that occur in women in the U.S. and 9% of cases in men, are diagnosed in never-smokers,” Kaufman said.

To a great extent, this is a function of straightforward math, said epidemiologist Ahmedin Jemal of the American Cancer Society. Fewer people smoke today than in previous decades — 15% in 2015, 25% in 1995, 30% in 1985, 42% in 1965. Simply because there are fewer smokers in the population, out of every 100 lung cancer patients, fewer will therefore be smokers. And that means more of them will be never-smokers.

There are also hints that the absolute incidence of lung cancer in never-smokers has been rising, said oncologist John Heymach of MD Anderson Cancer Center. Some data say it has, but other data say no. The stumbling block is that old datasets often don’t indicate a lung cancer patient’s smoking status, Heymach said, making it impossible to calculate what percent of never-smokers in past decades developed lung cancer.

Jemal, however, cautions that it is not the case that a never-smoker has a greater chance of developing lung cancer today than never-smokers did in the past.

Current cancer screening guidelines recommend a CT scan for anyone 50 to 80 years old who has smoked at least 20 pack years (the equivalent of one pack a day for 20 years, or two packs a day for 10 years, and so on) and who is still smoking or quit less than 15 years ago. Screening is not recommended for never-smokers because the costs of doing so are deemed greater than the benefits, Jemal said; thousands of never-smokers would have to be screened in any given year to find one lung cancer.

Still, low-dose CT can catch lung cancer in a significant number of never-smokers. A 2019 study in South Korea diagnosed lung cancer in 0.45% of never-smokers, compared to 0.86% of smokers. The researchers urged policymakers to “consider the value of using low-dose CT screening in the never-smoker population.”

“It used to be that the high-risk group” for whom CT screening is recommended “was the vast majority of lung cancer patients,” Heymach said. “But now that so many lung cancer cases are in nonsmokers, there is absolutely a need to reevaluate the screening criteria.”

Researchers are trying to improve screening by reducing the incidence of false positives — when CT finds lung nodules “or an old scar that you got 20 years ago,” he said. Those don’t pose a threat but have to be biopsied to ascertain that. Screening never-smokers would also be more efficient than it is today “if we could identify who, among nonsmokers, are at higher risk,” he said.

Cancer doctors already know part of the answer: women. Worldwide, 15% of male lung cancer patients are never-smokers. But fully half of female lung cancer patients never smoked. And women never-smokers are twice as likely to develop lung cancer as men who never put a cigarette to their lips.

Beyond sex, “nothing stands out as a single large risk factor that, if we only got rid of it, we would solve the problem” of lung cancer in never-smokers, said Josephine Feliciano, an oncologist at Johns Hopkins University School of Medicine. “But air pollution, radon, family history of lung cancer, [and] genetic predispositions” all play a role. Chronic lung infections and lung diseases such as chronic obstructive pulmonary disorder (COPD) also seem to increase risk.

None of those, with the possible exception of genetics and indoor pollution (cooking fires in some low-income countries), affect women more than men. So what’s going on?

At least one biotech believes that biological differences between lung cancer in never-smokers and smokers merits a new drug, and one that might be especially effective in women. “A different disease needs a different drug,” said co-founder and CEO Panna Sharma of Lantern Pharma. In fact Lantern, which is developing a drug for lung cancer in female never-smokers, believes that disease is so different it recently tried to convince the U.S. Food and Drug Administration to designate it an orphan disease, said Sharma.

Called LP-300, the Lantern drug increased overall survival from 13 months to more than 27, compared to chemotherapy alone, in female nonsmokers, in a small trial. It “targets molecular pathways that are more common in female nonsmokers than in any other group,” said Sharma, targeting the mutations EGFR, ALK, MET, and ROS1 (common in never-smokers) directly and boosting the efficacy of other drugs that attack them, such as erlotinib and crizotinib. Lantern plans a larger trial this year.

Smokers’ tumors tend to have more mutations overall, thanks to mutagen-packed cigarette smoke attacking their lungs, but scientists have developed more drugs for never-smokers’ lung tumors than for smokers’. For instance, EGFR and ALK mutations are more common in never-smokers. (Mandi Pike had the EGFR mutation, which was relatively fortunate: A drug targets it, and she has been cancer-free since November.)

The targeted drugs bollix up each mutation’s cancer-causing effects. KRAS mutations are more common in smokers’ lung tumors, and there are no KRAS drugs. (A KRAS drug for lung cancer is imminent, though, said thoracic oncologist Ben Creelan of Moffitt Cancer Center in Tampa, Fla.)

According to national guidelines, lung cancer in never-smokers should be treated the same as in smokers, said Creelan. “But I think we should reconsider this,” he said.

Because never-smokers have fewer tumor mutations, it’s harder to find them. So he said clinicians should be more aggressive about looking for actionable mutations in these patients. “I keep looking for a mutation until I find something important,” he said, adding that doctors might need better biopsy material or to use a different sequencing method in never-smokers.

In a cruel twist, the breakthrough drugs that take the brakes off immune cells, which then attack the tumor, are less effective in never-smokers’ lung cancer than in smokers’. The reason seems to be that smokers’ tumors have more mutations, said Mount Sinai’s Kaufman; the mutations often cause the tumor cells to have molecules on their surface that the immune system perceives as foreign and revs up to attack. Never-smokers’ tumors have few, if any, of those “come and get me” molecules. Immune cells therefore ignore them.

“In smokers, conversely, with more mutations, there is more for the immune system to recognize as bizarre and foreign, and so to provoke” an attack, Creelan said.

In contrast, never-smokers’ tumors are more likely to respond to targeted drugs, and as a result to be in remission for a long time or even cured. That’s because with fewer mutations, never-smokers’ tumors are more likely to have an “oncogene addiction,” Heymach explained: They are propelled by only one mutation. The plethora of mutations in smokers’ tumors means that there is usually a back-up cancer driver if a targeted drug eliminates cells with only one. “When a tumor has more and more mutations, blocking one is less likely to have an impact,” Heymach said. “But in nonsmokers, it can.”

Heymach called for more funding to study lung cancer in never-smokers. It “is an area that’s underserved and deserves more investment,” Heymach said. “It should be commensurate with the public health threat it represents.”

  • Thank you for this article! I can only hope more awareness and research is brought by it. I was a 44 year old, never smoker who’s world was shattered when diagnosed with lung cancer. The positive side is that it is one of the mutations referenced in this article, ALK, and I have been on targeted medication for 20 months now.
    Like other patients, it is hard to understand how this happened to me and I will most likely never know the cause. As a mother of three, I go through my day wondering what caused this to happen. Each time I take clothes out of the dryer, I wonder. As I remember working car loop as a schoolteacher, I wonder. As we use scented candles, the gas fireplace, our furnace, the a/c in my car that blows in my face, I wonder. When I see the dots of members on the global map of my support group, I wonder how this is so worldwide. More research and early screenings are imperative.
    Thank you, Sharon. Your article is a catalyst for Lung cancer support. May you Rest In Peace.

  • My wife (also a never-smoker) died last month of NSCLC; she blamed growing up in Cincinnati, but who knows? Exposed to second-hand smoke, cannot say about radon. And yes, the bulk of my reading indicates an absolute increase in incidence of lung cancer in never-smoker women.
    I had followed Sharon Begley’s writing back to her days at Newsweek–she is a major loss–R,I,P.

  • I’m very saddened by Sharon Begley’s passing as she was an outstanding scientific journalist whose reporting I found inspiring across decades.

    In 2011, the Fukushima nuclear disaster spread radioactive contaminants across the globe. The US Geological Survey measured fallout in the western US and US states, including East Coast states, reported radioactive iodine and cesium in precipitation. I personally strongly believe that Fukushima fallout helps explain the significant statistical uptick in lung cancers among never smokers beginning 2011. From the article above:

    “A 2017 study of 12,103 lung cancer patients in three representative U.S. hospitals found that never-smokers were 8% of the total from 1990 to 1995 but 14.9% from 2011 to 2013.”

    Radioactive contamination from Fukushima continues to this date.

    And then there is also the massive uptick in emissions from fracking, which include many carcinogenic chemicals.

    We are poisoning the environment and then blaming all the mass mortality events that follow on chance or some amorphous force, such as climate change, without analyzing the particulars.

    • While I can’t rule out the contribution of Fukushima, it does not explain that almost all the increased incidence has been among women.

  • One of the driving reasons for the increase is because there is a big increase in neuroendocrine tumors (carcinoid) that can be primarily in the lungs. Improvements in scanning technology and physician education have led to more cases of this once considered rare slow moving cancer being discovered. Even when present in the lungs this cancer is not related to smoking. Neuroendocrine cancer (carcinoid) is considered one of the most undetected cancers due to its slow growth rate and misdiagnosis. There are many different types of cancers that present in the lungs and there is not one that is caused only by smoking. Oncologist are well familiar with this. I guess to the layman it must come as a surprise that not all lung cancers are mostly from smoking. The best way to avoid all cancers is to eat and live well and don’t smoke.

  • What about secondhand smoke? Does proximity to major highways correlate or is that covered by air pollution?

  • Kudos to Sharon Begley for her lifetime of excellent reporting. In addition to drug discovery, we need better research into lung cancer triggers beyond smoking. A couple have been mentioned in the other comments.

    I believe another may be fungus/mold exposure, although not yet supported in the literature. My father, a nonsmoker, developed lung cancer in his 70s – within a year of prolonged exposure in a water-damaged condo that was not properly remediated. Fortunately, surgery was successful, the condo was fully remediated, and he even resumed jogging after recovery. In addition, some fungal masses in the lung mimic cancer, per

  • How many of them do you think were exposed to the dust from a building implosion? The respirable silica in the dust scars the lungs. It’s pH is the same as liquid Drano. The EPA made the connection between the two after researching the 4,000+ cases of cancer from the dust from the WTC collapse. Dr. Cate Jenkins was the lead researcher. Google it. It’s pretty interesting.

    • How about clay dust from cat litter? I started wearing an N95 mask when scooping the cat boxes a few years ago because of the dust. I don’t know whether it can cause cancer, but breathing that dust can’t be good for anyone.

  • Was this a sponsored article, or pitched via a third party media agency? If yes, please disclose. If not, apologies.

    • Great question. I would also like to know the answer to your question. I lost my 43 year old sister to lung cancer.

    • As stated at the very top of the article, STAT reporter Sharon Begley, the author of the article and a never-smoker, died recently from complications of non-small-cell lung cancer. The idea for it emerged, sadly, from her personal experience. That she worked on it despite living with this cancer and the side effects of treatment is a testament to her grit, her devotion to her craft, and the Sharon-ness that her colleagues knew and admired.

      Pat Skerrett

    • No. If you read the line at the top of the article, it was written because the author, Sharon Begley, was a never-smoker lung cancer patient herself who was actively battling the disease. She finished writing this article five days before she died, and it was published after her death.

    • I don’t know for sure but I doubt it was sponsored or similar. The author recently passed away from lung cancer, she was not a smoker. This was her final article she published before she died.

  • Sharon, thank you for an important article. Perhaps radon could get another look as a cause of lung cancer associated with extended home stay, such as during quarantine, which may have long-term effect in the future on lung cancers across the world.
    The link between radon and lung caner is well documented – 15,000 to 22,000 lung cancer deaths in the United States each year are related to radon. But I want to get attention on the antiquated standard and lack of testing for radon in the US. The EPA and the U.S. Surgeon General recommend that people not have exposures above 4 pCi/L on a long-term basis. But it is also estimated that a reduction of radon levels to below 2 pCi/L nationwide would likely reduce the yearly lung cancer deaths attributed to radon by 50%. However, even with an action level of 2.0 pCi/L, the cancer risk presented by radon gas is still hundreds of times greater than the risks allowed for carcinogens in our food and water. The 4pCi/L standard was set in 1986 but more recently WHO recommends mitigation above 2.7. Some comparisons of cancer risks illustrate that 4pCi/L is like smoking 8 cigarettes a day, 2.7 is closer to 5 cigarettes. So the goal should to have homes as close to zero radon exposure as possible.

    • Thank you Szymon for providing additional information. January is National Radon Action Month in Florida. I just ordered a radon test kit for our home.

Comments are closed.