Proton Radiotherapy for Cardiac and Pulmonary Vessel Sarcomas

Reviewer: Christine Hill-Kayser
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 28, 2010

Presenter: Dr. Yen-Lin Chen
Presenter's Affiliation: Massachusetts General Hospital, Radiation Oncology, Boston


  • Although cardiac and pulmonary vessel sarcomas are rare, they are not unheard of, and pose significant treatment conundrums.
  • The mainstay of treatment is surgical resection, but gross total resection is rarely achievable without causing potentially fatal damage to the heart and great vessels.
  • Radiotherapy and chemotherapy are usually part of the multidisciplinary approach to treatment of these tumors.
    • Local recurrence can be directly life-threatening, and local radiotherapy is often used in addition to surgery in either the neoadjuvant or adjuvant setting.
    • These tumors also have propensity to metastasize distantly; although they vary in their responsiveness to chemotherapy, systemic treatment may decrease the risk of development of distant disease.
  • The unique dosimetric properties of proton radiotherapy make it an appealing approach to these tumors. Protons may provide greatly increased conformality as compared to photons, and this may allow adequate delivery of dose to the target volume with relative sparing of nearby vital structures. In addition, use of proton therapy rather than intensity-modulated radiotherapy with photons (IMRT) allows reduction of the low-dose “bath” that may result from IMRT delivery. Reduction of this large low-dose volume may be particularly important where treatment of thoracic structures is concerned, as the volume of lung tissue receiving 5 and 20 Gray (Gy) has previously been identified as associated with risk of radiation pneumonitis (Allen, 2006), a potentially fatal complication.
  • In response to this dosimetric appeal, the authors performed an analysis of patients treated with proton radiotherapy for sarcomas of the heart and pulmonary vessels at their center, including clinical outcomes and toxicity.


  • Patient characteristics, radiation techniques, and outcomes were compared retrospectively.
  • Patients were treated between 2000 and 2009, and included 12 patients.
    • Eight patients were male and four were female.
    • Age at time of treatment ranged from 21 – 74 years (mean 49.2 years)
  • Primary tumor sites included:
    • Pulmonary artery/vein (4)
    • Right atrium (2)
    • Left atrium (4)
    • Interventricular septum (1)
    • Pericardium (1)
  • Histologies included:
    • Well-differentiated liposarcoma or fibrosarcoma (3)
    • High-grade intimal sarcoma (4)
    • Leiomyosarcoma (1)
    • Angiosarcoma (2)
    • High-grade undifferentiated spindle cell sarcoma (2)
  • For the first five patients, treatment planning was performed using pre- and postoperative chest CT and operative/surgeon reports.
    • Margins were based on echocardiogram to account for cardiac motion.
  • For the final seven patients, electrocardiogram-gated cardiac MRIs or 64-sliced high resolution cardiac CT scans were fused to planning 4D CT scans.
    • Cardiac internal target volumes (ITV) accounting for tumor motion through all 10 phases of a full cardiac cycle were then generated.
    • Patients were treated with respiratory gating at end expiration.
  • Median dose delivered was 66 Gy RBE (range 34.2 – 68.4; with the patient receiving 34.2 Gy RBE having also received 17.5 Gy via intraoperative radiotherapy).
  • Beam arrangements were chosen to minimize proton range perturbation by organ motion, and confirmed by range analysis using the Aqualyzer system.
  • Six of the 12 patients received chemotherapy as part of their treatment.


  • Median follow-up is 31 months.
  • Acute side effects included esophagitis, skin erythema, cough, and mild pericarditis.
  • No major cardiopulmonary events (radiation pneumonitis, myocardial infarction, heart failure, or arrhythmias) occurred.
  • Overall survival at 2 years was 75%, with three patients dead of disease.
    • One patient died of local failure, and two of distant metastatic disease.
  • Local control at 2 years was 83%, and distant metastasis-free survival was 50%.
    • Three patients were alive with distant metastasis (no local failure)
    • One patient had local failure and distant metastasis
    • Five patients had no evidence of disease

Authors' Conclusions

  • The authors conclude that proton radiotherapy appears to be a technically feasible and clinically well-tolerated local control modality for cardiac and pulmonary vessel sarcomas.
    • They note that the rate of distant metastasis remains high, but that high-dose proton radiotherapy may help reduce mortality from local failure.
  • They describe that electrocardiogram-gated cardiac imaging and respiratory gating can allow for treatment of these tumors while accounting for complex organ motion.

Clinical/Scientific Implications

  • The authors have developed a systematic approach to treating patients with these rare types of tumors that appears to be comprehensive and well thought out.
  • As evidence to this, they successfully treated 12 patients with rare thoracic sarcomas, none of whom encountered serious toxicity, including radiation pneumonitis.
    • The risk of radiation pneumonitis would be expected to be reasonably high with use of 3-D conformal or IMRT photon treatment.
    • The authors do not describe any patient who died from surgical or radiation complications.
  • Certainly, more patients in this cohort developed metastatic disease than local failure, and this may be demonstrative of a local control benefit with high-dose radiation as the authors suggest; however, this conclusion must be taken with caution given the small number of patients evaluated here, with relatively short follow-up.
  • Although the authors provided a good quantity of information regarding their treatment planning techniques, the information provided regarding treatment volume is relatively limited. Treatment volume delineation in sarcoma is somewhat controversial, as marginal treatment failures are common, and most practitioners feel that large treatment margins are necessary for treatment. More information on target delineation in the study population would certainly be of interest. Along the same lines, further information on types of local failure (in-field versus marginal) might help treatment decisions in the future.
  • In sum, the authors report on important data here, which may serve to be hypothesis-generating. While further data are awaited, the information presented here provides preliminary evidence that proton radiotherapy may be appropriate treatment for these challenging patients, as well as description of a feasible technique that accounts for organ motion.