Definitive high dose photon/proton radiotherapy for unresected chordomas

Reviewer: Geoffrey Geiger MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 14, 2009

Presenter: Y-L. Chen, W. Kobayashi, S. Childs, A. Rosenberg, G.P. Nielsen, D. Rosenthal, H. Suit, T. DeLaney
Presenter's Affiliation: Massachusetts General Hospital, Boston, USA
Type of Session: Scientific


  • Chordomas are rare and slow-growing but locally aggressive malignancies arising from embryonic notochordal elements.
  • Areas most frequently affected include the skull base and sacral regions.
    • In adults, 50% involve the sacrococcygeal region, 35 percent occur at the base of the skull near the sphenooccipital area, and 15 percent are found in the vertebral column (Heffelfinger, 1973).
  • Aggressive initial therapy has been found to improve overall outcome and typically involves surgery, which serves the dual purpose of providing tissue for histopathological diagnosis and tumor debulking.
    • Due to the proximity to critical anatomical structures (e.g., spinal cord, sacral nerve roots), treatment is challenging, and complete resection (R0) is often not possible, necessitating postoperative radiation.
    • Unfortunately, by the time patients become symptomatic, tumors are frequently so large that they are not amenable to complete surgical resection (Magrini, 1992).
  • Conventional photon irradiation results in poor outcomes in patients with gross residual disease following surgical resection.
    • In a series published in 1996 by Catton et. al., 31 patients were treated with conventionally fractionated radiation to a median dose of 50 Gy in 25 fractions over 5 weeks (range 25-60 Gy).
      • Median survival was 62 months (range 4-240 months) from diagnosis.
      • One complete and no partial responses were identified in 23 assessable patients.
      • The local control rate was 27%, although 85% achieved useful and prolonged palliation of pain.
  • Even with multimodality therapy, local recurrence remains the most common failure pattern, resulting in an adverse overall survival.
  • In the past, prior to 3D treatment planning and photon-based intensity modulated radiation therapy (IMRT), doses capable of rendering local control were not attainable without unacceptable toxicity.
  • Charged particle beam therapy with either proton or carbon ion therapy allows the delivery of higher doses of conformal radiation due to the dose characteristics of proton beam radiotherapy, including finite range and steep dose fall off beyond the Bragg peak.
    • Protons can achieve an approximately 10–15% dose decrease per millimeter in the fall-off region.
  • An approach using a combination of photon- and proton-based radiotherapy has been utilized at Massachusetts General Hospital (MGH) since 1973 (Hug, 1995) for skull base and spinal cord/paraseptal chordomas.
  • In the largest series of patients treated with charged particle therapy, 195 patients with resected chordoma of the skull base or cervical spine were treated with a combination of photon and proton radiation (Debus, 1997) using prescribed target doses ranging from 63 to 72.9 cobalt gray equivalents (CGE).
    • At a median follow-up of 54 months, 69% were relapse-free. Five- and 10-year progression-free survival (PFS) rates were 70 and 45%, respectively.
  • A phase II study of high-dose photon/proton radiotherapy in the management of spine sarcomas was published in 2009 (DeLaney, 2009).
    • In a subset of unresectable patients, local control rates were found to be high, which prompted the investigators to retrospectively evaluate the question of whether patients can be treated safely and effectively with radiation alone.

Materials and Methods

  • This abstract represents a retrospective review of 19 patients selected from Massachusetts General Hospital in Boston, MA.
  • All patients had tumors judged unresectable by the referring surgeon.
    • Two patients had cervical chordomas, 1 was thoracic, 2 were lumbar, and 15 were sacral, of which 12 were involving the S1 or S2 nerve roots.
    • 11 patients underwent core (n = 10) or incisional biopsy (n = 1).
      • Morphologic analysis still represents the basis of pathologic diagnosis for chordomas.
  • All patients were treated with definitive proton-based radiotherapy from 1997 to 2008.
  • All patients had newly diagnosed disease at the time of radiotherapy.
  • For protons, kilovoltage orthogonal and individual field port imaging was performed daily with any requisite table adjustments and reimaging where necessary to verify accuracy prior to each fraction.
  • The RBE for proton radiation was set at 1.1. Thus, the dose unit, Gy RBE, was the physical dose in Gy x 1.1.


  • Median age was 67 years (range 30-82).
  • Mean total dose was 77.3 Gy RBE (range 76.6-77.4 Gy RBE), delivered using primarily protons (mean dose 46.8 Gy RBE) with a component of 3D conformal or intensity-modulated photons (mean dose 30.6 Gy).
  • At a mean follow-up of 45 months (range 12-136 months, minimum 1 year), 10 of 11 patients had local control.
    • One progressed locally at 2 years. One developed metastatic disease at 1 year and died at 15 months but had local control.
  • Overall, few late effects were observed. They included moderate fibrosis in the soft tissues overlying the spine in most, incomplete voiding and erectile dysfunction in one patient, and sacral insufficiency fracture in another.
  • Overall, 18 of 19 patients demonstrated the absence of local progression.
    • One patient suffered from local failure, although his primary tumor was larger than 10 cm.
    • Another patient suffered a distant failure after being treated for a 17 cm primary tumor, but had intact local control at the time of the diagnosis of his metastatic disease.

Author's Conclusions

  • Because of their anatomical location, biology, and high locoregional recurrence rates, chordomas are tumors that are infrequently amenable to R0 resections and frequently require adjuvant radiation.
  • Gross total resection is desirable but limited by the close proximity of these tumors to brainstem, cranial nerves, or critical blood vessels.
  • Proton therapy is an attractive treatment option affording the ability for excellent local control with limited acute and long-term toxicities.
  • In this retrospective review, high-dose proton-based radiotherapy can be an effective treatment for patients with unresectable chordomas and offers a chance for preservation of nerve function with minimal side effects.
    • More than 90% of patients in this retrospective cohort demonstrated local control at a median follow-up of 45 months.

Clinical/Scientific Implications

  • In this retrospective review, DeLaney et. al., report data from a relatively small cohort of patients with unresectable chordomas, and demonstrated excellent local control following treatment using primarily protons with a component of 3D conformal or intensity-modulated photons.
  • A much larger cohort of patients (n = 290) treated with combined photon/ proton radiotherapy at MGH with doses ranging from 60-68 CGE has been analyzed and demonstrated far lower rates of local control (64%) (Munzenrider, 1999) following resection.
    • In that study, median follow-up was 41 months, but was extended to almost 22 years; however, doses were lower and patients were treated with more rudimentary treatment planning techniques.
  • The data presented here demonstrate that delivery of adequate dose to unresectable chordomas is feasible and tolerable with a combination of photons and protons, achieving excellent rates of local control warranting continued study with reports including longer follow-up.
  • Furthermore, this study suggested that definitive treatment with radiation is not only highly effective vis-à-vis local control, but also feasible with preservation of neurological function with relatively few side effects.
  • The data presented here appear interesting and promising, with potentially improved results as compared to historical data; however, the retrospective nature of the study, small sample size, and relatively short follow-up time presented preclude making statements regarding effectiveness in the definitive setting.

From the National Cancer Institute