Analysis of V20 and Radiation Pneumonitis on SWOG 0023: A Phase III Trial of Concurrent Chemoradiation and Docetaxel Consolidation in Stage III Non-small Cell Lung Cancer

Reviewer: Charles Wood, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 8, 2006

Presenter: L.E. Gaspar
Presenter's Affiliation: University of Colorado at Denver Health Sciences Center
Type of Session: Scientific


  • SWOG 0023 was a prospective, randomized, multi-institutional phase III trial evaluating the use of external beam radiotherapy with concurrent cisplatin/etoposide chemotherapy, followed by consolidation docetaxel chemotherapy, and then maintenance with gefitinib versus placebo in patients with inoperable stage IIIA or stage IIIB non-small cell lung cancer.
  • V20 (% total lung volume receiving ≥20 Gy) has previously been found to correlate with the development of pneumonitis.
  • This analysis represents a subset of enrolled patients that have been evaluated for correlation of the incidence, timing, and severity of pneumonitis with a number of factors, including V20 and FEV1.

Materials and Methods

  • The radiation records of 440 out of the 620 accrued patients have thus far been reviewed by both QARC and the primary investigator (LG). Patient- and treatment-related characteristics were found to be similar between this subset and the entire study population.
  • The radiotherapy regimen consisted of a total dose of 61 Gy, delivered in 1.8 - 2 Gy fractions; CT simulation was required, dose inhomogeneity corrections were not utilized, and dose-volume histograms were submitted for tumor target volumes, the spinal cord, and the bilateral lungs.
  • Study patients were required to have either an FEV1 = 2 liters or a total lung capacity = 800 cc in the contralateral lung, as determined by quantitative V/Q scan.
  • A V20 of <35% was recommended but not required for patients enrolled in the trial.
  • Pneumonitis was defined by the National Cancer Institute Common Toxicity Criteria version 2.0.
  • Pneumonitis data was available for 494 patients starting chemoradiation, 343 patients starting consolidation chemotherapy, 100 patients randomized to the gefitinib arm, and 109 patients randomized to the placebo arm.


  • The study was closed prematurely on 4/15/05, after an unplanned interim analysis of the data demonstrated no benefit for the gefitinib arm
  • The median survival for all patients was 19 months, and there was no statistically significant difference between the 2 arms.
  • Of all patients beginning treatment in the study, 7% developed grade 3 pneumonitis at some point during the treatment course.
  • The time-course incidence of development grade 3 pneumonitis was as follows:
    • <1% during concurrent chemoradiation
    • 7% during consolidation chemotherapy
    • <1% during maintenance gefitinib/placebo
  • Of the 444 patients for whom radiation details are known, V20 was found to exceed 35% in 18% of patients.
  • There was no significant association between V20 and patient gender, age, stage (IIIA vs. IIIB), or FEV1.
  • V20 was found to be significantly associated with the subsequent development of grade 3 pneumonitis (p=0.005).
  • V20 > 35% was found to be significantly associated with the subsequent development of grade 3 esophagitis (p=0.032).
  • Median survival for patients with V20 > 35% was found to be significantly worse versus those with a lower V20 (12 vs. 24 months, p=0.0006).

Author's Conclusions

  • The frequency of grade 3 pneumonitis (7%) observed in this trial was within reasonable limits and correlated with V20 > 35%.
  • It is not possible to ascertain whether the decreased survival in patients with higher V20 was due to tumor bulk or treatment-related toxicity
  • Future SWOG studies will mandate that patients have V20 < 35%

Clinical/Scientific Implications

V20 is being increasingly utilized as a treatment planning parameter in an effort to decrease the incidence of radiation-induced pneumonitis. However, as the measure is quantitative and not qualitative, it does not provide information on the "quality" of the lung receiving radiation. The volume of lung able to tolerate a given amount of radiation may certainly differ between that of a "never-smoker" and that of a patient with a long-standing history of heavy tobacco use. A measure or tool that would assist in estimating the allowable V20 for an individual patient, based on the "quality" of that patient's lungs, would be very useful.

Additionally, caution should be exercised when drawing conclusions from the finding that V20 was significantly associated with decreased survival. As pointed out by the authors, no efforts have been made thus far to control for tumor size, and tumor size alone would be expected to impact on both survival and on the volume of lung receiving significant dose.