Treatment of NSCLC patients with proton beam-based stereotactic body radiation therapy: A dosimetric comparison with photon plans highlights the importance of range uncertainty

Reporter: Arpi Thukral, MD MPH
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 13, 2011

Presenter/Author: Seco, J.
Presenter's Affiliation:Massachusetts General Hospital


  • Non-small cell lung cancer accounts for 80% of all lung cancers.
  • Surgery is the gold standard and main curative modality for early-stage lung-cancers (T1-T2, N0) and results in favorable treatment outcomes.
  • However, stereotactic body irradiation (SBRT) is an increasingly popular and widely accepted treatment modality for patients with early-stage NSCLC, especially those that are inoperable. Local control rates at 3 years using photon SBRT are in the 85% range (Timmerman, et. al.).
  • Photon SBRT patients are typically treated with 14-18 Gy per fraction for a total of 3-5 fractions, with a 3D conformal technique using 8-10 beams. Normal tissue sparing is required, as well as image guidance.
  • Recently, proton radiation therapy has been proposed for use in SBRT in lung cancer. It is thought that the potential advantage of proton therapy is its potential to minimize dose to adjacent normal tissues, especially chest wall and contralateral lung, given its high conformality index and decreased exit dose due to its characteristic Bragg Peak.
  • This study was performed to analyze how range uncertainties for protons may impact its role for SBRT in a small cohort of lung cancer patients.

Materials and Methods

  • 10 patients with early–stage NSCLC were treated at Massachusetts General Hospital with proton therapy based SBRT, using a passive scattering technique with 2-3 beams.
  • Patient characteristics:
    • Most patients had upper lobe tumors, and all patients’ tumors were required to have <5 mm motion.
    • Median age: 72.7 years
  • Separate plans using photon SBRT were made for each patient and then compared.
  • Photon treatment:
    • PTV= 5mm axial and 10 mm longitudinal to GTV30
    • 5-6 coplanar and 3-4 non coplanar beams
    • 42-48 Gy in 3-5 fractions was prescribed and normalized to 70-80%
  • Proton treatment:
    • 3-5 beams
    • PTV= 2mm range uncertainty and 4-8 mm for lateral margin for proximal and distal
    • Protons volumes took into account a margin for 3.5% range + 2mm to account for range uncertainties. This translated to a 4-8mm extra margin distally and proximally.
    • Proton plans were normalized to the 95% isodose line (as compared to photon plans, normalized to the 70-80% isodose line).
  • Conformatlity (CI) and heterogeneity indices (HI) were calculated for both sets of plans.
    • CI= V95%Rx/VPTV (refers to the how well the treated volume conforms to the PTV target volume)
    • HI=Dmax/DRx (refers to how homogeneous the dose distribution is; lower HI=more homogeneous, less hot spots)
  • High and low-dose regions within the treatment volume were defined. High-dose regions were those volumes receiving more than 50% of dose and low-dose regions were those receiving less than 50% of dose.
  • Based on these dose regions, dosimetric comparisons were made between the photon and proton plans.
    • The 50% mark was chosen because it reflects the inflection point where dose fall-off occurs.


  • In high-dose regions, the average volume receiving: >95% of the prescription dose was greater for proton than for photon SBRT
    • 46. 5 cc vs. 33.5 cc (p=0.009), respectively
    • This was likely due to range uncertainty margin given for proton volumes.
    • Proton HI =1.04 vs. photon HI = 1.24
    • Proton CI= 2.46 vs. photon CI= 1.56.
  • For tumors in close proximity (within 8 mm) to the chest wall, the volume of the chest wall receiving 30 Gy or more (V30) was 5-7 cc larger for protons than for photons (p=0.06), respectively.
    • However, for those patients with tumors > 8 mm from the chest wall, no difference in volume of chest wall irradiated was seen (p=0.3).
  • For low-dose regions, maximum esophagus dose and lung V5 were calculated. They were found to be smaller for protons than for photons (p=0.019 and p<0.001, respectively).

Author’s Conclusions

  • Protons can create larger treatment volumes in high-dose regions when compared to photons because of range uncertainties.
  • In turn, this can result in increased dose (close to prescription dose) to nearby organs, such as chest wall.

Clinical Implications

  • This well-designed study of 10 patients offers an interesting overview of the advantages and disadvantages of proton therapy in the setting of SBRT for early-stage lung cancer patients.
  • Advantages of protons in this setting include: low HI (very homogeneous dose distribution) due to prescription to 95% isodose line as opposed to 60-90% for photon plans. Although this is an advantage, photon plans can often generate higher dose to the central part of the tumor, which may be the hypoxic area and benefit from dose escalation. Therefore, the high HI seen with photon therapy, may actually be advantageous in lung tumors.
  • The other main advantage of protons is the sparing of normal tissues in low-dose regions, especially considering contralateral lung and esophagus.
  • Disadvantages of protons include increased treatment volume because of range uncertainties. SBRT was adapted as a strategy for lung tumors in order to provide high dose to tumor, while sparing normal lung. If treatment volume needs to be increased due to margin for range uncertainty with protons, the advantage is lost. Increased high-dose treatment volume can cause unacceptably high doses to adjacent normal tissues, especially near the mediastinum and chest wall.
  • Furthermore, proton therapy is very sensitive to motion, which cannot always be controlled in lung cancer treatment.
  • SBRT using protons for early-stage lung cancer is feasible, but needs to be used in a very carefully selected group of patients. Risks to critical organs are serious, and patients with tumors close to the chest wall may not be optimal candidates for proton therapy.
  • Physicians should be aware of this issue and should not undertake this type of treatment, without proper immobilization and image guidance, physics support, and comparison of proton plans to photons plans for each individual patient.
  • SBRT using protons should be studied in a larger scale prior to being adapted into clinical practice.

Frequently Asked Questions