Proton Therapy for Early-Stage Lung Cancer at Loma Linda University Medical Center

Reviewer: Christine Hill-Kayser, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 13, 2009

Presenter: D. Bush
Presenter's Affiliation: Loma Linda University Medical Center
Type of Session: Scientific


  • Non-small cell lung cancer (NSCLC) remains the most frequent cause of cancer-related death in the United States, and is a prominent cause of death worldwide.
  • Survival correlates with disease stage at diagnosis, and patients with early-stage NSCLC may have five-year survival rates approaching 80%.
  • Surgical resection remains a mainstay of treatment for early stage disease, and more aggressive surgical treatment consisting of lobectomy has been demonstrated to provide superior outcomes to more limited surgeries, such as wedge resection or segmentectomy (Ginsberg, 1995).
  • Unfortunately, a significant proportion of patients with early-stage NSCLC have co-morbidities that limit the safety of surgical approaches; such patients may be deemed "medically inoperable."
  • Such patients may be treated with radiotherapy alone. Historically, use of photon radiation has provided worse outcomes when compared to surgical resection; however, it is utilized in cases when surgery is not deemed to be safe.
  • When conformal radiotherapy is utilized, dose may be limited by constraints to other normal tissues, including the normal lung tissue, spinal cord, heart, esophagus, and great vessels.
  • Due to physical aspects of proton beams, proton radiotherapy may allow delivery of more conformal treatment; proton delivery results in a Bragg peak, or region where most dose is delivered, with very little exit dose. These properties may allow avoidance of other normal structures and may lead to improved local control and survival.
  • Although dose benefits have been demonstrated in other sites, use of proton radiotherapy to treat lung tumors may be complicated by several qualities of lung tumors and tissues.
    • Because lung tissue consists of a significant portion of air, proton stopping power may be more uncertain when compared to other tissues.
    • As a result, risk exists for Bragg peak overshoot. If this occurs, it will lead to large dose deposition in an erroneous region.
  • Despite these concerns, significant potential advantages exist to use of proton therapy for treatment of lung cancers. These include decreased dose to the contralateral lung, spine, and mediastinal structures, which may allow tumor dose escalation.
  • The below trial was undertaken to explore the feasibility of use of proton radiotherapy in treatment of medically inoperable NSCLC.

Materials and Methods

  • The trial described here was the first trial examining the role of proton radiotherapy in treatment of lung cancer at Loma Linda University, and was initiated in 1996.
  • The trial was designed as a phase I feasibility study.
  • Eligible patients had clinical stage I medically inoperable NSCLC.
    • A total of 83 patients were accrued, with median follow-up of 48 months.
  • Initial dose delivery consisted of 51 CGE delivered in 10 fractions. After initial feasibility analysis, dose was increased to 60 CGE delivered in 10 fractions.
  • Treatment planning was performed in a manner similar to that used for conformal photon radiation planning:
    • All target delineation was performed using lung windowing.
    • PET/CT scan was used to delineate tumor versus other soft tissues.
    • 4-dimensional planning was used to capture free-breathing target motion.
    • Treatment planning was performed using the Odyssey system, with beam apertures cut to encompass the entire internal target volume, taking into account target motion.
    • Lateral and oblique beams were preferentially utilized, with weighting as indicated.
    • Normal structures were delineated, with dose-volume histograms generated.


  • A total of 83 patients were enrolled on this study.
    • Median age at diagnosis was 73.8 years.
    • All patients had stage I disease; 41 had tumor classification 1 (T1, no more than 3 cm) tumors, and 42 had T2 (3 – 6 cm) tumors.
    • Median tumor size was 3.7 cm (range 1.1 – 8.9 cm)
    • Median forced expiratory volume in 1 second (FEV1) was 1.1L (43% of predicted).
  • Median follow-up was 48 months.
  • Treatment plans achieved excellent ITV coverage, with minimal dose to other normal structures. Of note, the mean lung volume receiving 20 CGE (V20) remained less than 5%.
  • The study's primary endpoint was local control (LC), which approached 70% in the overall study population.
    • When patients were considered by tumor classification, LC was 88% for patients with T1 tumors, versus 50% for those with T2 tumors.
  • Overall survival (OS) at five years was less than 20%.
  • Patients were observed to die often of co-morbid causes, and disease specific survival (DSS) was thus considered; DSS at 5 years approached 70%.
  • The Charlson Co-morbidity Index was used to further investigate DSS. Using this tool, predicted overall survival based on co-morbidities alone was calculated for the study population. The predicted death rate correlated with the actual death rate from co-morbid conditions in the study population with 95% confidence.
  • The authors observed a 3.6% rate of clinical pneumonitis requiring steroid treatment (n = 3). A 3.6% risk of rib fracture (n = 3) was also observed. Lung function was noted to remain relatively stable over five years of follow-up.

Author's Conclusions

  • The authors conclude that, based on the data presented here, there is rationale for the study of proton radiotherapy in treatment of lung cancers.
  • They note that the trial described here supports feasibility of further trials, and that further trials are needed. They mention hope that collaboration between institutions may allow such trials to be undertaken expeditiously.

Clinical/Scientific Implications

  • The authors present an interesting study, which has important implications in the treatment of NSCLC.
  • Although surgical treatments remain the mainstay of treatment for early-stage NSCLC when feasible, such treatments are not without risk. Possibilities for outcomes with use of radiotherapy that approach those achieved with surgery certainly have important implications for patient care.
  • Definitive, conformal photon treatment with standard fractionation has been demonstrated to yield local control rates of 50-70%, compared to 90% with definitive surgery (Bradley, 2005; Sibley, 1998; Ginsberg, 1995).
  • Use of other radiotherapy techniques, such as stereotactic radiosurgery, may provide improved outcomes without use of surgery when compared to conventional radiation.
  • Based on the data presented here, proton radiotherapy may also represent a possible means of achieving excellent outcomes using definitive radiation to treat early-stage NSCLC.
  • Having said this, multiple other factors must be taken into account:
    • First, the cost and limited accessibility of proton therapy will, for the time being at least, limit its use. Availability and cost-effectiveness of photon-based stereotactic radiosurgery may prove to be superior to those of protons if clinical outcomes are similar.
    • As the authors point out, significant uncertainty regarding proton dosimetry in lung tissue exists. Although this feasibility study supports safe use of protons to treat lung tumors, with minimal resultant toxicity, further data will be needed for eventual assessment of appropriate use of proton radiation in lung.
  • As a feasibility study, this trial represents an important contribution that may have long-term implications for patient care. As the authors note, further study is certainly indicated for further understanding of both efficacy and cost-effectiveness.