STOP SMOKING IN EARLY LUNG CANCER

January 27, 2009 — "It is never too late for people to stop [smoking], even when they have lung cancer," proclaims an editorial published online January 21 in the British Medical Journal.

The editorial accompanies a meta-analysis that provides "preliminary evidence that smoking cessation after diagnosis of early-stage lung cancer improves prognostic outcomes," according to its authors.

The adjusted estimates suggest that the risk for death is halved in patients who stop smoking, say the researchers, led by Amanda Parsons, research fellow at the UK Center for Tobacco Control Studies at the University of Birmingham, United Kingdom.

"The estimated number of deaths prevented is larger than would be expected from a reduction of cardiorespiratory deaths after smoking cessation, so most of the mortality gain is likely to be due to reduced cancer progression," they write.

"These findings indicate that offering smoking-cessation treatment to patients presenting with early-stage lung cancer may be beneficial," they conclude.

The difference in survival between patients who stopped smoking and those who continued is "striking," said H. Jack West, MD, medical oncologist at the Swedish Cancer Institute in Seattle, Washington, and author of the Blowing smoke blog on Medscape Oncology. This effect is larger than that seen with postoperative chemotherapy, he added.

"If smoking cessation can improve true cure rates even modestly, it is absolutely an intervention worth pursuing," he said.

However, both the study authors and Dr. West emphasized that the finding pertains only to early-stage lung cancer patients.

Does Not Pertain to Advanced Disease

The situation is quite different for patients with advanced disease — and they form the majority of patients diagnosed with lung cancer.

Most patients diagnosed with lung cancer are in the last months of their lives, note editorialists Tom Treasure, MD, MS, professor of cardiothoracic surgery at University College Hospital in London, and Janet Treasure, MD, PhD, professor of psychiatry at King's College in London, United Kingdom, who are married.

"Fewer than 1 in 3 patients with lung cancer survive even 1 year," they add.

Some healthcare professionals might balk at discussing smoking cessation with this patient population, suggesting that it "is inhuman to dwell on the matter" because it "adds to feelings of guilt and takes away a life long comfort from the dying patient," they write in the editorial.

Elaborating further to Medscape Oncology, Dr. Tom Treasure, who was until recently chair of the National Confidential Enquiry into Patient Outcome and Death, said: "I get rather frustrated with people tormenting patients who are near to death, trying to jolly them along by saying we can give you some more chemotherapy. Is that really the best thing?"

"All patients know they should stop smoking, and we should make it clear to individuals that we approve this general message," Dr. Tom Treasure said.

"Getting down to the detail, we should emphasize that smoking cessation matters to most, perhaps nearly all, patients," he said. "They are entitled to know that it can make a difference."

However, he cautioned, "this general advice should be tempered with humanity."

Reduction in Risk for Death

The meta-analysis looked at 10 observational studies, 5 in patients with nonsmall-cell lung cancer (NSCLC; n = 860) and 5 in patients with small-cell lung cancer (SCLC; n = 1069). In 9 of the 10 studies, most of the patients were diagnosed as having early lung cancer, the researchers note.

Continued smoking was associated with a significantly increased risk for all-cause mortality (hazard ratio [HR], 2.94; 95% confidence interval [CI], 1.15 - 7.54) and recurrence (HR, 1.86; 95% CI, 1.01 - 3.41) in early-stage NSCLC, and for all-cause mortality (HR, 1.86; 95% CI, 1.33 - 2.59), development of a second primary tumor (HR, 4.31; 95% CI, 1.09 - 16.98), and recurrence (HR, 1.26; 95% CI, 1.06 - 1.50) in limited-stage SCLS.

Life-table modeling on the basis of these data estimated that, for early-stage NSCLC, the 5-year survival for 65-year-old patients was 33% if they continued to smoke and 70% if they stopped smoking. For limited-stage SCLC, these estimates were 29% for those who continued to smoke and 63% for those who quit.

This is the first study to have estimated the effect of smoking cessation on prognosis after a diagnosis of lung cancer, note the researchers. "Although unadjusted estimates suggest that the associated increase in risk of continuing (or the benefit of cessation) is modest, at around 20%, the adjusted estimates suggest a more than doubling in the risk of death from continued smoking," they add.

"These estimated effects are large," the editorialists note. "Patients and those caring for them should be given this information because the potential benefit is great," they explain.

In an interview with Medscape Oncology, Dr. Tom Treasure wondered about the magnitude of this effect on survival, and whether it was likely. "In such a cancer, where there have been years of exposure, it does seem a little strange that suddenly stopping smoking can have such a large effect," he said.

Need for Proper Context

The editorialists note that this new information "is valuable," but they add that "its application may be limited" because of the generally poor prognosis for most patients with lung cancer.

This is one of the issues with these latest findings — the results come from studies in which patients were mainly diagnosed and treated in the early stages of lung cancer — whereas in real life, most patients (75% to 80%) with lung cancer are diagnosed with advanced disease, as lead researcher Ms. Parsons acknowledged to Medscape Oncology.

"We are not suggesting" referral for smoking cessation for patients with advanced disease and poor prognosis, she emphasized.

However, for the 15% to 20% of patients who are diagnosed early enough to receive curative treatment (stages 1 to 3a NSCLC and limited-stage SCLC), the new findings "indicate that there may be a meaningful difference in survival when you compare those who quit with those who continue to smoke after diagnosis."

The figures that the British researchers report "do make sense in the context of nationally reported survival rates," Ms. Parsons said. For example, Cancer Research UK cites the following 5-year survival rates for early-stage NSCLC: 43% to 73% for stage 1, 25% to 46% for stage 2, and 19% to 24% for stage 3a disease.

"Our review found that a mixture of stage 1 to 3a NSCLC patients who smoked had a 5-year survival of 33%," she noted; quitting smoking increased this to 70%.

Ms. Parsons was responding to some of the criticism leveled at her teams' findings by the American Council on Science and Health (ACSH). In its newsletter, Gilbert Ross, MD, medical director at the ACSH, said the survival rates "are staggeringly high" and that they appear to be "outlandishly optimistic." He admitted that "we haven't analyzed the science of it . . . so we can't condemn the study outright, but these figures seem unlikely."

Ms. Parsons emphasized that the figures need to be taken in their proper context — i.e., just early-stage lung cancer patients rather than lung cancer patients as a whole.

Stopping Smoking in Early-Stage Patients

Ms. Parson noted that, because the studies were observational, causality cannot be assumed, but the review provides "preliminary evidence that stopping smoking may be beneficial in early-stage lung cancer patients."

"I hope that doctors will read the article and, in conjunction with their clinical judgment, decide whether it would be important to support their early-stage patients to stop on an individual basis," she told Medscape Oncology.

However, before smoking cessation for early-stage lung cancer can become part of standard practice, it needs to be tested in a large-scale randomized controlled trial, she explained. "Until then, data are encouraging that survival benefits can be seen in this group, and it is worthwhile encouraging patients to stop while assessing each case on its individual characteristics."

Approached for independent comment, Dr. West said that the new findings are "very promising and interesting, particularly as we compare the magnitude of benefit with approaches like chemotherapy."

"One central shortcoming of this work is the heterogeneity of patients studied, including those with NSCLC, SCLC, and advanced-stage disease, along with the majority who had early-stage NSCLC or limited-stage SCLC. Another is that this work is observational, and there were imbalances in the demographics of patients who quit [and those who] continued smoking in some studies," he said.

Therefore, he said, "we cannot state conclusively that the improved survival in patients who quit smoking was because of this change in behavior, but the magnitude of the survival difference, as well as the fact that these results were seen in such a broad range of patients with lung cancer, is striking."

"Given that these benefits of smoking cessation may eclipse the magnitude of such interventions as postoperative chemotherapy, it only adds to the limited information previously available to support quitting," Dr. West stated.

However, Dr. West also emphasized that these new findings apply only to patients with early-stage disease who are potentially curable.

The situation can be quite different in patients with advanced disease who are being treated with palliative intent. Here, he emphasized, "it is especially important to acknowledge the powerful addiction to nicotine that many patients have, and to respect that some may weigh the challenge of smoking cessation as having too negative an effect on their quality of life to justify a potential survival benefit."

"Nevertheless, chemotherapy and other systemic treatments that are the cornerstone of treatment for advanced lung cancer also entail a careful balance of quality of life with potential survival benefit; for many patients, smoking cessation may be associated with an improvement in pulmonary symptoms," he added.

"Overall, it is important to recognize that smoking is a charged issue in the lung cancer community. Lung cancer is reflexively associated with smoking, and many patients justifiably feel an element of 'blame the victim' from others and even themselves," Dr. West said.

"We need to approach counseling for smoking cessation constructively and without judgment," he concluded.

Ms. Parsons reports receiving reimbursement from Pfizer for attending a conference, and one of her coauthors, P. Aveyard, from the UK Centre for Tobacco Control Studies, reports carrying out consultancy work for the manufacturer of a smoking-cessation aid.

BMJ. 2010;340:b5569, b5630. Published online January 21, 2010. Abstract, Abstract