Ultralow-dose CT bests X-Ray for lung cancer follow-up, but is it enough?

MINNEAPOLIS – Minimal-dose computed tomography was superior to chest radiographs for surveillance after curative lung cancer resection in a randomized controlled trial involving over 300 patients.

"Minimal-dose CT should be the modality of choice for surveillance after resection of lung cancer," Dr. Waël Hanna said at the annual meeting of the American Association for Thoracic Surgery.

^{Copyright Martin Allred}

Dr Waël Hanna

Repeated radiation exposure and a high false-positive rate have been stumbling blocks to national lung cancer screening with low-dose spiral CT, despite the technology demonstrating 20% fewer lung cancer deaths compared with chest x-ray in asymptomatic heavy smokers in the National Lung Cancer Screening Trial (NLST). Minimal-dose CT of the chest delivers a radiation dose of 0.2 mSv per scan, which is comparable to chest x-ray at 0.16 mSv and lower than a diagnostic CT or low-dose CT at roughly 8 mSv and 1.5 mSv, he said.

The 311 patients in the current study were prospectively enrolled after curative resection and underwent minimal-dose CT and chest x-ray at 3, 6, 12, 18, 24, 36, 48, and 60 months. A total of 1,137 pairs of chest X-ray and CT scans were analyzed by radiologists blinded to the other modality.

Minimal-dose CT detected 94.2% of the new or recurrent lung cancer, compared with 21.1% for chest x-ray (P value .0001), said Dr. Hanna, a thoracic surgery fellow at the University of Toronto.

The increased sensitivity came at a cost of significantly lower specificity (86% vs. 99.9%) and positive predictive value (25.1% vs. 91.6%; both P less than .0001). The negative predictive value for minimal-dose CT, however, was almost perfect (99.7% vs. 96.1%; P = .007).

More importantly, of the 63 patients diagnosed with new or recurrent cancer, 49 (78%) had asymptomatic disease detected only on minimal dose CT, Dr. Hanna said. Two-thirds of the asymptomatic patients were diagnosed within the first year of surveillance and 94% within 2 years of initial surgery.

"Why is this important? Because when you find it at an earlier stage, earlier in time, you can do something about it," he said.

Asymptomatic patients who were restaged and given curative surgery or radiation went on to live a median of 69 months (range, 12-76) after the initial operation, compared with a median survival of 25 months (range, 6-48) among asymptomatic patients given palliative treatment after restaging (P less than .001).

The 14 patients with symptomatic recurrent or new cancer had a median survival of only 15 months (range, 7-63) with palliative care.

"We are not saying that minimal-dose CT improves survival because these two patient populations are different," Dr. Hanna said. "The patient who presents with asymptomatic disease of the chest and is a candidate for surgery is clearly different from the patient who presents with brain metastases and is symptomatic. But follow-up with minimal-dose CT allows us to identify this cohort of patients in whom close surveillance after surgery is not futile, in whom close surveillance after surgery is amenable to intervention and treatment, and is associated with long survival."

Several prominent guidelines, including those from the AATS and National Comprehensive Cancer Network, have moved to include regular CT scans in the wake of the NLST, but evidence is lacking to suggest that earlier treatment of recurrence leads to better outcomes.

Invited discussant Dr. Michael Jaklitsch of Brigham and Women’s Hospital, Boston, said that the investigators were able to take a group of patients who would have had a 5-year survival of 50% and raise it to 75% through aggressive surveillance and show that they were "truly curing" these patients.

"Is this enough data to change our personal practices today," he posited. "For me personally, the answer is yes. This single paper presents me with enough data to say I will use minimal-dose CT scan as my sole method of screening for recurrence of early-stage lung cancer moving forward."

Dr. Jaklitsch questioned whether there were subpopulations in whom minimal-dose CT would not work, such as the obese or those with surgical clips. Dr. Hanna said that radiologists at his center are more comfortable using low-dose CT for surveillance because of the risk of scatter in either of these subgroups or in those with mediastinal involvement.

Dr. Hanna also noted that minimal-dose CT is not available everywhere, but Dr. Jaklitsch said that he took the specifications from the paper to his community hospital and they said they could be done. "So at least in the U.S., this will have dramatic penetrance," he added.

^{Copyright Martin Allred}

Dr. David Sugarbaker