Traditional versus up-front 18-FDG PET staging of non-small cell lung cancer (NSCLC): A Dutch Co-operative randomized study

Reviewer: Neha Vapiwala, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 6, 2004

Presenter: G.J. Herder
Presenter's Affiliation: POORT Study Group, Netherlands
Type of Session: Scientific

The use of FDG-PET scan as an add-on study in non-small-cell-lung-cancer (NSCLC) patients scheduled to undergo surgery has been shown to be a cost-effective and useful tool in guiding patient management. When employed after traditional staging studies, just prior to the planned surgical intervention, PET has revealed previously unsuspected extrathoracic disease and changed the management plan, preventing unnecessary surgery. This study evaluated the use of PET scan as an up-front study, performed upon initial presentation, in lieu of more traditional staging procedures. Its goal was to evaluate if initial PET scanning could simplify the staging process by providing comprehensive and accurate whole-body data equivalent to the information obtained from multiple traditional staging studies. As the appropriate and timely management of NSCLC patients is so strongly dependent on the patient's stage, any simplification of the process is worthy of investigation.

Materials and Methods

  • From 9/1999 to 6/2001, 465 patients were enrolled by 23 hospitals in the Netherlands 
  • Patient cohort represented 22% of all NSCLC patients registered in the national database
  • All patients were required to have suspected NSCLC based on history, physical examination, chest x-ray, bronchoscopy
  • Eligible patients could not have overt stage III/IV disease or diabetes mellitus
  • They were randomized upon first presentation to undergo traditional staging workup (TWU, n= 233) or PET scan (PWU, n=232)
  • Patients were stratified by ECOG performance status and institution of enrollment
  • All baseline dempgraphics were equally balanced in both arms, including age and gender
  • Patients randomized to TWU were managed according to international practice guidelines on patient management
  • Patients randomized to PWU received only a PET scan, with treatment algorithm as follows:
  •     PET positive lesions for distant metastases (DM) or lymph node involvement (LNI) required pathologic verification and CT or MRI scans of the brain, as PET is recognized to be inferior for CNS evaluation 
  •     PET scans negative for DM or LNI but with PET-avid, non-central tumors were recommended to undergo thoracotomy 
  •      PET scans with FDG-coin lesions with non-diagnostic bronchoscopy and presumed to be benign were followed closely.
  • Primary endpoints of the study were the effect of PET on the number of invasive and non-invasive staging studies, the duration of the staging process, the accuracy of the staging (agreement between clinical and pathologic findings), and the cost efficacy as  compared to traditonal workup.  
  • The study was designed to detect a 30% reduction in the number of required staging tests in the PWU group compared to the TWU.


  • There was no statistically significant difference between the average number of staging tests performed in the TWU vs. the PWU arms (7.88 vs. 7.90).
  • There was no statistically significant difference between the clinical staging accuracy between the two arms (TWU 0.85 vs. PWU 0.78, p= 0.073).  
  • The time period to final staging was 23 days in the TWU group and 14 days in the PWU group.
  • There was no siginificant difference noted between the two arms in cost (TWU 1,156 Euros vs. PWU 1,924 Euros).


Author's Conclusions

  • The use of initial FDG-PET early in the staging process of suspected NSCLC did not significantly reduce the number of staging tests.
  • The use of PET provided data of similar accuracy to the traditional tests.
  • There was no significant cost advantage to the use of PET vs. typical staging studies.
  • The use of PET did reduce the overall duration of the staging process by 9 days compared to traditional workup.
  • PET up-front has similar overall accuracy in the staging of NSCLC as traditional staging studied, with no siginificant impact on test substitution.

Clinical/Scientific Implications

The data presented here offer some interesting insight on the use of FDG-PET as an initial part of staging NSCLC patients. Whereas the role of PET as an additonal tool following completion of standard imaging and other work-up tests has been demonstrated to be a cost-effective tool that can prevent unnecessary surgical intervention, the data here are not as encouraging for PET as a substitution tool. Although its use provided data of comparable accuracy to more traditional workup, there was no siginificant benefit in terms of reduction in cost or number of studies. However, PET did help to shorten the staging process by 9 days, which is important for the expeditious treatment of NSCLC patients.

Of note, most staging studies and procedures in the Netherlands are ordered by the pulmonologists and oncologists, often after conferreing with each other on each patient case. Given the long enrollment period, the issue is raised of changing practice patterns and increasing expertise on the part of radiologists in radiographic study interpretation, and the possible effect of this on the study data. However, the authors state that clinical practice trends did not change significantly over the time period of the study, thereby reducing the effect of this potential confounder. Nonetheless, it is interesting to note that there was only a 20% staging error with the traditonal workup in this trial, markedly better than the typical 40% error rate reported previously with traditional imaging studies. Regardless, the PET scan was able to meet this standard of accuracy. Future studies may further explore the use of PET as a complementary staging tool rather than a substitutive one. 

Oncolink's ASCO Coverage made possible by an unrestricted Educational Grant from Bristol-Myers Squibb Oncology.