Phase II clinical studies of high-dose proton therapy in stage I/II and III non-small cell lung cancer

Reporter: Gita Suneja, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 14, 2011

Presenter: Joe Y. Chang, MD PhD
Presenter's Affiliation: MD Anderson Cancer Center


  • NSCLC is a prevalent and deadly disease. Over 160,000 cases are diagnosed per year in the U.S., and over 157,000 deaths occur.
  • For early stage disease, surgery is the mainstay of therapy; however, limited surgery (or wedge resection) is not adequate even for early stage disease. More substantial surgery such as lobectomy or pneumonectomy is required for adequate disease control (LCSG 821, 1995). For advanced disease, chemoradiation with or without surgery is the standard.
  • Some patients with early stage or low-risk disease are not ideal candidates for the major surgical procedures required. In recent years, new treatment options have emerged for these patients, such as stereotactic body radiation therapy (SBRT), which have outcomes comparable to surgical resection.
  • With 3-dimensional radiation therapy, median OS for NSCLC was approximately 1.5 years and 5-year overall survival was about 20%. Based on data obtained from the SEER database, conventional radiation therapy offers a median 5-7 month survival benefit., but no benefit in 5-year overall survival, suggesting delay of progression of disease rather than cure.
  • Conventional doses used in 2D and 3D planning were 60-66 Gy.
  • More recently, data from Memorial Sloan Kettering and Washington University suggest that dose escalation to doses > 80 Gy can extend overall survival, perhaps even curing NSCLC. However, the benefits of increased dose must be balanced with increased acute and long-term toxicity.
  • Over the past several years, Dr. Chang and his colleagues at MDACC have been studying IMRT and proton therapy as a means to escalate dose, while protecting the heart, spinal cord, and contralateral lung. In a previous study published in 2006, proton therapy reduced normal tissue dose compared with 3DCRT and IMRT in stage III NSCLC patients.
  • This purpose of this study was to evaluate the efficacy and toxicity of dose escalation with proton therapy in patients with Stage I-III NSCLC.

Materials and Methods

  • 31 patients with medically inoperable T1-3 N0 M0 NSCLC were treated with proton therapy. Non-centrally located T1 tumors were excluded. These patients were treated to 87.5 Gy (RBE) in 2.5 Gy (RBE) per fraction.
  • 54 patients with stage III NSCLC were treated proton therapy and concurrent weekly carboplatin and palcitaxel. These patients received 74 Gy (RBE).
  • All patients underwent 4D-CT based proton treatment planning with periodic re-simulation for adaptive re-planning.
  • PET scans were obtained before treatment for diagnosis, as well as after treatment to monitor treatment response
  • Primary outcomes: local control, overall survival, and toxicity


  • In the cohort receiving 87.5 Gy (RBE):
    • Median follow-up was 13 months (range 3.2 to 53.8 months)
    • Grade 3 dermatitis was seen in 17% of patients
    • None experienced grade 4 or 5 toxicity
    • Local failure within the PTV was 9.7%, regional nodal failure was 6.5%, and distant metastasis occurred in 16.1%.
  • In the cohort receiving 74 Gy (RBE) with concurrent chemotherapy:
    • Median follow-up was 19.5 months (range 3.2 to 47.1 months)
    • Grade 3 toxicities included dermatitis (9%), esophagitis (11%), esophageal stricture (3%), pneumonitis (3%), and pulmonary fistula (1%)
    • Local failure rate was 16.7%, regional nodal failure was 7.4%, and distant metastasis occurred in 40.7%.
    • Median overall survival was 29.4 months. 1-year overall survival and progression-free survival were 86% and 63% respectively.
  • Most common failure pattern was distant.
  • 11 of these patients underwent adaptive radiation therapy planning. The patients in this cohort tended to have larger tumor size, which decreased in density during the course of treatment, leading to overdose of the contralateral lung. As compared to IMRT, proton is more sensitive to changes in target volume.
  • Additionally, PET SUV was found to predict for recurrence and survival in pts with stage III NSCLC. Pre-treatment and post-treatment PET scans were obtained with corresponding SUV 1 and SUV 2. An SUV2 > 3.6 was found to predict for local recurrence.

Author’s Conclusions

  • Proton therapy for NSCLC reduces toxicity and allows further dose escalation.
  • Adaptive planning is indicated in select patients.
  • Post-treatment PET SUV2 predicts for local recurrence.

Clinical Implications

  • This study contributes to the growing body of literature supporting the use of proton therapy in NSCLC.
  • The local control rates achieved here and in other more modern studies are encouraging, and suggest that proton therapy rivals surgery in terms of local control outcomes.
  • The lack of adverse outcomes and normal tissue toxicity are also promising.
  • In spite of the success of dose escalation to prevent local recurrence, metastasis to distant sites continues to be the more common cause of mortality. As we improve and fine tune local therapy, we also need to consider better treatments to address the problem of distant disease.
  • The implications of post-treatment PET in predicting local recurrence are profound. Perhaps patients with a high probability for local recurrence based on post-treatment PET will be candidates for further dose escalation or other additional treatments.
  • More research, specifically prospective randomized controlled trials, are needed to determine the utility of proton therapy in NSCLC.