Chordoma: long-term follow-up after radical photon irradiation

To retrospectively analyze the long term results of treatment and the patterns of failure for patients with chordoma of the sacrum, base of skull and mobile spine treated predominantly with postoperative photon irradiation.

Forty-eight adult patients with chordoma of the sacrum (23), base of skull (20) or mobile spine (5), were seen between 1958-1992. Forty-four were referred post operatively with overt disease and 31 of these were irradiated with conventionally fractionated radiation to a median dose of 50 Gy/25 fractions/5 weeks (range 25-60 Gy). Eight received a hyperfractionation protocol of 1 Gy, 4 hourly, 4 times a day (median 40 Gy/44 fractions/14 days), two sacral patients were treated with a hypofractionation protocol and three cases with skull base tumours were referred elsewhere for proton therapy. Endpoints measured were survival from diagnosis, objective response rate, symptomatic response rate and clinical or radiological progression-free survival from radiotherapy.

Median survival was 62 months (range 4-240 month) from diagnosis with no difference between clival and non-clival presentations. One complete and no partial responses were identified in 23 assessable patients. A subjective response was recorded for 12/14 (85%) with pain and 10/23 (45%) with neurological signs or symptoms, and the median time to progression for those with overt disease was 35 months (range 5-220 months). There was no survival advantage to patients receiving radiation doses > 50 Gy (median 60 Gy) compared to doses < 50 Gy (median 40 Gy). There was no difference between the conventional or hyperfractionation regimens with respect to the degree or duration of symptomatic response, or in progression-free survival. Fourteen patients who progressed after irradiation were retreated with surgery (6), irradiation (7) or both modalities (1). Median survival after retreatment was 18 months, and the only two symptomatic responses seen were with reirradiation, and after failure of relatively low dose initial therapy. At last follow-up, 35 were dead of or with disease, seven are alive with disease, and two are disease-free. Thirty-eight had local disease persistence as the sole site of failure, and four developed distant metastases initially or subsequently.

Overt residual chordoma is rarely cured with conventional external beam irradiation, but treatment does provide useful and prolonged palliation of pain for most patients. Chordoma is a disease with low metastatic potential, and better local control may improve survival. Complete resection rates may be improved for patients with sacral disease by using planned excisions in centres experienced in treating this rare disease. Because radiation therapy may prove to be more successful in controlling microscopic disease, it should be considered as a pre- or postoperative adjuvant to a macroscopically complete resection. Patients with skull base disease should also be resected in centres specializing in this surgery, but complete excision is unlikely. These patients will not obtain local control with conventional photon irradiation, and suitable patients should be considered for irradiation with stereotactic photon or particle beam therapy. For patients who progress after irradiation, there is limited symptomatic benefit to retreatment with surgery or reirradiation, and this should be limited to treating life-threatening complications.