Over the lasat decade, PET scans have become commonplace in the staging of NSCLC. There's an older post that reviews the concept of PET scans in providing metabolic imaging, as well as a podcast that provides a more complete discussion of PET scanning in oncology, with a focus on lung cancer.

A recent paper in the New England Journal of Medicine highlights the main benefit seen with PET scans for lung cancer staging. This particular paper looked at the newer combination PET/CT scans that have become very common and largely replaced separate PET scanners in many parts of the world. PET/CT scanners now allow us to see the superimposed images that provide very good detail of the shape and size of internal parts of the body (the CT portion) along with the metabolic uptake of these areas (the PET portion).

This study was conducted in Denmark and randomized 189 patients in the process of staging for possible resectable NSCLC to either conventional staging with CT-based imaging and mediastinoscopy (required) or the same treatment with a PET/CT scan. They were looking for a significant difference in the frequency of "futile thoracotomies", so basically trying to see if PET scans cut the frequency of people undergoing a major lung surgery for no benefit. Their definition of this was rather liberal, since it included not only patients who had higher stage disease than they intended to pursue surgery on (stage IIIA with N2 nodes, stage IIIB, or stage IV), those with an unsuspected benign cause, an exploratory surgery (which I imagine would have been to determine what the cause of the lesion was), or someone who had a recurrence within a year of their surgery. This last point is a controversial one, because we might think after the surgery that a person undergoing surgery should have gotten it if their staging after surgery wasn't too high -- but if they recurred very quickly, it's clear that these patients didn't benefit.

The concept of a futile thoracotomy is a useful one, because we might rush off to the operating room and be happy to get the tumor out, but if a cencer recurs before a patient has recovered, that wasn't a favor. Just today, in fact, I saw a patient for a second opinion who had a right pneumonectomy (entire right lung removed, which is a major loss, since the right lung is bigger than the left one because the heart sits on the left side of the chest), but this man developed a recurrence just a few months after surgery. Now he's getting chemo and struggling to manage with a pleural effusion around his one remaining lung.

So getting to surgery is a great goal if you really have a fair chance of benefiting, but it's not something that's of value if you sneak your way in but really won't benefit. The measure of "futile thoracotomies" captures the proportion of people receiving surgery that probably wasn't in their best interests at all.

This study found that people randomized to PET/CT scans in addition to conventional staging were more likely to be found inoperable (38 vs 18), while there was a significant reduction in the number of futile thoracotomies in the PET-CT recipients (21 vs. 38, p = 0.05). Interestingly, the real difference in futile thoracotomies was from patients who had a recurrence in less than a year after surgery, which is the most bold definition.

There's little doubt that PET scans for staging lung cancer are here to stay. More people may be offered the "opportunity" for surgery if we follow a "don't ask, don't tell" policy and just don't look very hard for advanced disease, but it really doesn't do the patient a favor to have them undergo a major surgery that wouldn't end up providing a meaningful chance of benefit.