Low-Dose Spiral Computed Tomography Screening For Lung Cancer: Not Ready for Prime Time

Edward F. Patz Jr. and Philip C. Goodman

Department of Radiology, Duke University Medical Center, Durham, North Carolina

	ARTICLE

TOP ARTICLE REFERENCES

Over the past several decades, lung cancer screening has been promoted as a strategy to decrease the number of deaths from this disease. The success of these screening programs has been based on the premise that if a tumor were detected early enough, improvements in outcome would follow. This theory was initially tested using chest radiographs and sputum cytology as the screening modalities. Multiple randomized trials were initiated, and preliminary results appeared promising (1- 3). These studies consistently showed that patients in the screened group had earlier stage disease, more resectable tumors, and an increase in survival time (time from diagnosis to cancer death). However, these data proved misleading, as both short- and long-term follow-up revealed no statistically significant reduction in lung cancer deaths, the ultimate goal of a diagnostic screening program (2, 4). A number of interpretations have been proposed to explain these somewhat conflicting results, but needless to say there are no data to prove that screening for lung cancer with chest radiographs or sputum cytology reduces mortality, and no medical organization currently recommends lung cancer screening.

Now, low-dose spiral computed tomography (CT) is being suggested as a new screening technique. Some investigators have promoted CT because it can detect smaller, and presumably earlier stage, tumors than chest radiographs. The assumption is that that a 5- to 10-mm nodule found on CT is more curable than a 2- to 3-cm lesion seen on chest radiographs. If this were proved to be true then the smaller lesions should consistently represent localized, early-stage disease unlike larger tumors. It is unknown, however, whether this reduction in size of radiologically detectable lesions corresponds to the critical period during which tumors metastasize and thus advance the stage at presentation.

In fact, by the time a tumor reaches 5 mm in diameter it has already undergone approximately 20 doubling times and represents almost 10⁸ cells (5). Thus even those lesions considered small by radiologic standards are late in the natural history of this disease. In regard to this, several recent reports have examined the relationship between tumor size and survival and stage at presentation. One study of 510 patients found no statistical relationship between tumors  3 cm and survival; patients with 3-cm masses had the same outcome as patients with 1-cm nodules (6). In a related study of 620 patients, there was no statistically significant relationship between the size of the primary tumor and stage at presentation. Patients with a 1-cm primary tumor presented a similar stage distribution as those with 2- to 3-cm tumors (7). These studies suggest that the biologic behavior of tumors (i.e., potential to metastasize) is variable and may be determined before lesions can be detected radiographically. Development of metastases may be more dependent on factors such as tumor genetics or angiogenesis than size.

Although multiple experimental and clinical studies continue to support or refute lung cancer screening, the only true way to test its validity is by appropriate hypothesis-driven trials. Over the past few years several groups have initiated screening programs, but most are nonrandomized studies (8- 11). To date only prevalence screen data from these trials have been published. The results confirmed several expected findings: CT was more sensitive for detecting lung nodules than conventional chest radiographs, some of these nodules were lung cancer (up to 27 per 1,000), and most of the lung cancers (up to 85%) were early-stage, resectable tumors.

Again these initial results appear encouraging, but there are still no mortality data, and several other findings have become apparent. First, CT was so sensitive that in one study 51% of participants had nodules detected at the prevalence screen (11). This does not include "missed" nodules found on the Year 1 incidence CT examination and then retrospectively confirmed to be present on the prevalence screen study. The vast majority of these nodules will most certainly be benign, but establishing the etiology may incur morbidity from percutaneous needle biopsy or thoracotomy, radiation exposure from follow-up CT studies, and participant anxiety. Second, although up to 85% of tumors detected by CT were Stage I disease, there was in fact no "stage shift." For a true stage shift there must be not only an increase in early-stage disease, but a concomitant decrease in late-stage disease. A comparison of the low-dose CT data with the results of previous chest radiographic trials revealed that the number of patients with advanced stage disease was the same in both (about 3 per 1,000 patients). Third, the cost of a screening program including the required follow-up studies of the innumerable indeterminate nodules would be staggering. If the proposed 40 million smokers and 40 million nonsmokers were screened, some have now estimated the cost at more than $70 billion/yr. Health care dollars are not unlimited. How to judiciously use these resources (e.g., primary prevention, screening, and treatment) has yet to be determined.

Screening for lung cancer is a complex, confusing, and controversial subject. At this time no screening trial has shown a reduction in lung cancer mortality, and we should not adopt unproven practices without the appropriate confirmatory data. Although CT prevalence screen results have received much publicity and some project that CT screening will "save lives," it is premature to implement guidelines or set standards on the basis of predicted outcomes. We strongly encourage individuals to participate in ongoing trials so that the appropriate answers can be ascertained, but believe that currently mass CT screening for lung cancer is "not ready for prime time."

	References

TOP ARTICLE REFERENCES

1. Frost JK, Ball WC, Levin ML, Tockman MS, Baker RR, Carter D, Eggleston JC, Erozan YS, Gupta PK, Khouri NF, et al . . Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Johns Hopkins Study. Am Rev Respir Dis 1984; 130: 549-554 [Medline].

2. Fontana RS, Sanderson DR, Woolner LB, Taylor WF, Miller WE, Muhm JR, Bernatz PE, Payne YS, Pairolero PC, Bergstralh EJ. Screening for lung cancer: a critique of the Mayo Lung Project. Cancer 1991; 67: 1155-1164 [Medline].

3. Flehinger BJ, Melamed MR, Zama MB, Heelan RT, Perchick WB, Martini N. Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Memorial Sloan-Kettering study. Am Rev Respir Dis 1984; 130: 555-560 [Medline].

4. Marcus PM, Bergstralh EJ, Fagerstrom RM, Williams DE, Fontana R, Taylor WF, Prorok PC. Lung cancer mortality in the Mayo Lung Project: impact of extended follow-up. J Natl Cancer Inst 2000; 92: 1308-1315 [Abstract/Free Full Text].

5. DeVita VT, Young RC, Canellos GP. Combination versus single agent chemotherapy: review of basis for selection of drug treatment of cancer. Cancer 1975; 35: 98-110 [Medline].

6. Patz EF Jr,, Rossi S, Harpole DH Jr,, Herndon JE, Goodman PC. Correlation of tumor size and survival in patients with stage IA non-small cell lung cancer. Chest 2000; 117: 1568-1571 [Abstract/Free Full Text].

7. Heyneman LE, Herndon JE, Goodman PC, Patz EF Jr. Stage distribution in patients with a small (< 3 cm) primary non-small cell lung cancer: implication for lung cancer screening. (Submitted for publication)

8. Sone S, Takashima S, Li F, Yang Z, Honda T, Maruyama Y, Hasegawa M, Yamanda T, Kubo K, Hanamura K, Asakura K. Mass screening for lung cancer with mobile spiral computed tomography scanning. Lancet 1998; 351: 1242-1245 [Medline].

9. Kaneko M, Eguchi K, Ohmatsu H, Kakinuma R, Naruke T, Suemasu K, Moriyama N. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 1996; 201: 798-802 [Abstract/Free Full Text].

10. Henschke CI, McCauley DI, Yankelevitz DF, Naidich DP, McGuinness G, Miettinen OS, Libby DM, Pasmantier MW, Koizumi J, Altorki MK, Smith JP. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999; 354: 99-105 [Medline].

11. Swensen S. Lung cancer screening: low-dose CT [abstract]. Society of Thoracic Radiology, March 2000, San Diego, CA.