|Home | About | Journals | Submit | Contact Us | Français|
We previously reported a phase 2 trial of 69 patients with newly diagnosed anaplastic or aggressive oligodendroglioma who were treated with intensive procarbazine, CCNU (lomustine), and vincristine (PCV) followed by high-dose thiotepa with autologous stem cell rescue. This report summarizes the long-term follow-up of the cohort of 39 patients who received high-dose thiotepa with autologous stem cell support. Thirty-nine patients with a median age of 43 (range, 18–67) and a median KPS of 100 (range, 70–100) were treated. Surviving patients now have a median follow-up of 80.5 months (range, 44–142). The median progression-free survival is 78 months, and median overall survival has not been reached. Eighteen patients (46%) have relapsed. Neither histology nor prior low-grade oligodendroglioma correlated with risk of relapse. Persistent nonenhancing tumor at transplant was identified in our initial report as a significant risk factor for relapse; however, long-term follow-up has not confirmed this finding. Long-term neurotoxicity has developed only in those patients whose disease relapsed and required additional therapy; no patient in continuous remission has developed a delayed neurologic injury. This treatment strategy affords long-term disease control to a subset of patients with newly diagnosed anaplastic oligodendroglioma without evidence of delayed neurotoxicity or myelodysplasia.
Oligodendroglioma is a chemosensitive brain tumor. In contrast to other types of malignant glioma, many anaplastic oligodendrogliomas respond dramatically and completely to chemotherapy (Cairncross and Macdonald, 1991; Cairncross et al., 1994; Chamberlain and Kormanik, 1997; Friedman et al., 1998). Symptomatic or enlarging low-grade oligodendrogliomas (aggressive oligodendrogliomas) and anaplastic mixed oligoastrocytomas are also chemosensitive (Kim et al., 1996; Mason et al., 1996). Radiotherapy is known to be effective for oligodendroglial neoplasms, but the optimal timing is controversial (Gannett et al., 1994). Radiotherapy can have serious long-term neurotoxic effects, particularly when the treatment field is large. These effects are progressive and of greatest concern in patients with expected long-term survival. Oligodendrogliomas, including anaplastic oligodendrogliomas, may have an indolent course, making patients with these tumors most vulnerable to delayed radiotherapy-induced neurotoxicity.
In 2003, we reported the results of a multicenter phase 2 study of 69 patients with newly diagnosed anaplastic oligodendroglioma who were treated with intensive procarbazine, CCNU (1-[2-chloroethyl]-3-cyclohexyl-1-nitrosourea; lomustine), and vincristine (PCV)4 followed by high-dose thiotepa with autologous hematopoietic stem cell rescue in chemosensitive patients; radiotherapy was deferred until relapse (Abrey et al., 2003). This report summarizes the long-term follow-up of those 39 patients who completed the planned therapy.
Between 1993 and 2000, 39 patients with newly diagnosed anaplastic oligodendroglioma were treated in a prospective phase 2 trial of high-dose chemotherapy with autologous stem cell rescue; the details of this study have been reported previously (Abrey et al., 2003). This prospective multicenter study was open to any patient with a newly diagnosed anaplastic or aggressive oligodendroglioma or anaplastic mixed oligoastrocytoma. All patients had to have pathologic confirmation of an oligodendroglial neoplasm, but the requirement for anaplasia could be waived, provided that a lesion demonstrated clinical or radiographic evidence of aggression. Clinical aggression was defined as neurologic deterioration, increasing seizure activity, or increasing steroid requirements. Radiographic aggression was defined as contrast enhancement and tumor enlargement on MRI or CT. Anaplastic mixed tumors were required to have 25% oligodendroglial elements. All pathology was reviewed centrally by one of the authors (D.A.R.). All patients signed written informed consent, and the protocol was approved by the Institutional Review Board at each participating institution.
Induction chemotherapy consisted of three or four cycles of intensive PCV, chosen to maximize dose intensity during induction therapy. Cycles were administered every six weeks as follows: lomustine (CCNU), 130 mg/m2 p.o., was given on day 1; vincristine, 1.4 mg/m2 (no cap on dose), was given intravenously on days 8 and 29; and procarbazine, 75 mg/m2 per day p.o., was given from day 8 to day 21. Response to treatment was evaluated with serial neuroimaging, and tumor size was measured as the maximum cross-sectional diameter of the enhancing mass. Complete response (CR) was defined as the absence of all enhancing tumor, with the patient off corticosteroids and neurologically stable. Partial response (PR) was defined as 75% reduction in tumor size in a neurologically stable or improving patient on stable or decreasing doses of corticosteroids.
For the high-dose therapy, a total dose of 900 mg/m2 of thiotepa was administered as 150 mg/m2 intravenously every 12 h over 72 h for a total of six doses. Peripheral stem cells were reinfused four days (96 h) after completing thiotepa infusion. Standard supportive care measures were used for all patients. Follow-up clinical, laboratory, and imaging evaluations were performed every three months for two years, every six months for three years, and then yearly or as dictated by symptoms.
The primary goal of this report is to summarize the long-term follow-up of the 39 patients who received high-dose thiotepa with autologous stem cell support. The other 30 patients who left the study prior to high-dose thiotepa are not included in this analysis or report. The Kaplan-Meier (1958) product limit method was used to analyze survival data. Time to progression was calculated from date of study entry to the day of radiographic progression or last follow-up. Overall survival was calculated from day of study entry to date of death or last follow-up. Discrete variables were analyzed by the chi-squared method with Yates correction for expected values of less than 5.
Thirty-nine patients (23 men and 16 women) completed the induction and transplant regimens (Table 1). Their median age was 43 (range, 18–67) and median KPS was 100 (range, 70–100). Twelve patients had a history of prior low-grade oligodendroglioma, 29 had pathologic confirmation of anaplastic oligodendroglioma, and 10 had either a mixed oligoastrocytoma or an aggressive oligodendroglioma. Twenty-two had a complete tumor resection prior to starting chemotherapy, and 17 had only a partial resection.
Of the 39 patients who underwent transplant, 18 (46%) developed progressive disease by a median of 33.5 months (range, 10–92) after transplant. Neither histology (oligodendroglioma vs. mixed oligoastrocytoma, P = 0.75) nor a prior diagnosis of low-grade oligodendroglioma (P = 0.81) correlated with relapse. Patients with an initial complete resection had the same risk of relapse as those with a partial resection (P = 0.87). Evidence of residual nonenhancing tumor was identified as a significant risk factor for relapse in our initial report, but with prolonged follow-up, this variable no longer reaches statistical significance (P = 0.09).
Treatment at relapse varied, but most patients received cranial radiotherapy with or without salvage chemotherapy, with a median survival of 19 months (range, 3–73+). Eight of the 18 patients treated at relapse remain alive, with durable disease control ranging from 11 to 73 months. Most of these eight patients have some degree of permanent neurologic disability as a result of either their tumor or tumor-related treatment. One patient died of delayed neurologic complications of therapy, and another surviving patient has significant secondary disability.
Median survival of transplanted patients has not been reached, with a median follow-up of the 29 surviving patients of 80.5 months (Fig. 1); 10 patients have died. Estimated median progression-free survival is 78 months. The quality of long-term survival appears to be excellent for those patients in a continuous remission; all have a KPS in the range of 90 to 100, and the majority are working at or close to the level of their prior occupation. Only one patient is on disability related to her tumor. No patient has developed progressive neurologic dysfunction that could be attributed to the therapy. No delayed myelodysplastic syndromes have been detected.
At the time of our initial report, the status of chromosomes 1p and 19q was not available. Since that time, we have obtained information on 10 of the 39 patients (Table 2). Six of the 10 patients had loss of heterozygosity (LOH) on both 1p and 19q. Only one patient had intact 1p and 19q; the other three had either 1p or 19q LOH. The small number of patients and retrospective nature of this information preclude drawing significant conclusions; however, it is noteworthy that two patients with 1p and 19q LOH had early progression and poor response to salvage therapy, whereas one patient with intact 1p and 19q had excellent initial disease control and has enjoyed prolonged survival after relapse.
Combined treatment with intensive PCV followed by high-dose chemotherapy with autologous stem cell support achieved excellent disease control and preservation of neurologic function in this cohort of patients with anaplastic and aggressive oligodendrogliomas. At present, there is a considerable variety of treatment options for patients with newly diagnosed anaplastic oligodendroglioma. Many patients are treated with focal radiotherapy alone, and the results of a recently reported Radiation Therapy Oncology Group (RTOG) study would support this strategy (Cairncross et al., 2004). The chemosensitivity of oligodendrogliomas has persuaded many practitioners to offer chemotherapy alone, usually temozolomide or PCV, and defer radiotherapy until relapse; several reported series support this approach; however, other reports suggest that chemotherapy alone is inadequate to provide long-term disease control (Chibbaro et al., 2004; Paleologos et al., 1999; Soffietti, 2004). This protocol was designed to replace radiotherapy with a myeloablative dose of thiotepa in an effort to eradicate microscopic disease and provide nonneurotoxic antitumor therapy. With long-term follow-up, more than half of the patients remain in remission without any evidence of treatment-related neurologic injury.
Less than half of the patients who completed the planned therapy developed tumor progression. Relapse rarely occurred within the first year of follow-up but has been observed as late as seven years after completion of therapy, which indicates the need for long-term follow-up. Furthermore, most patients (83%) whose disease relapsed responded to salvage therapy with control of their recurrent tumor for six months or longer. It would appear that both radiotherapy and additional salvage chemotherapy, including temozolomide, could induce further disease control.
Surviving patients in a continuous CR are functioning at a high level that is at or close to their level of function prior to tumor diagnosis. Detailed neurocognitive testing and quality-of-life measures were not included in this protocol, and it is therefore possible that subtle cognitive deficits or reduced quality of life were not detected. However, no clinically apparent neurologic deterioration or treatment-induced myelodysplasia have developed.
Combined allelic loss of chromosomes 1p and 19q are associated with durable responses to chemotherapy in oligodendroglial neoplasms (Cairncross et al., 1998; Ino et al., 2001; Smith et al., 2000). Our study did not analyze the molecular genetic status of our patient’s tumors prospectively; however, since our initial report, we obtained genetic information regarding the molecular status on 10 of our patients. The available data suggest that most of our patients had 1p and/or 19q LOH and excellent response to therapy; however, several with combined allelic loss had a brief response to chemotherapy and varying duration of survival after salvage therapy. Furthermore, one patient with intact 1p and 19q had a durable response to initial chemotherapy and prolonged survival following salvage radiotherapy. This highlights the difficulty of predicting an individual patient’s treatment response based on prognostic genetic factors and emphasizes the importance of analyzing the prognostic implications of 1p and 19q status prospectively in order to develop appropriate therapeutic guidelines. Recent publications suggest that other genetic alterations play an important role, either complimenting or negatively impacting the effect of 1p and 19q LOH (Ino et al., 2001). Other cases of excellent outcome in patients with intact 1p and 19q further indicate that other genetic alterations are likely important prognostic markers of response to therapy and overall outcome (Ino et al., 2001).
These encouraging data suggest that some high-grade oligodendrogliomas may be treatable by chemotherapy alone and that effective treatment can result in excellent neurologic function for many years. These results should be interpreted in the context of the relatively small sample size, young age, and excellent performance status of our patients. Careful selection of patients who are medically appropriate to be considered for high-dose chemotherapy with autologous stem cell support introduces a selection bias that may be reflected in the long-term outcome of our patients. However, this is the group of patients who may derive the most benefit from an aggressive treatment strategy that defers radiotherapy in an effort to avoid delayed cognitive impairment.
The authors thank their many data managers and research nurses: Martin Kleber, Lynette Brunaldi, Marilyn David, Jolene Lewis, Rebecca Desgroseilliers, Pat Lada, and Pina Giuliano. The authors thank Judy Lampron for her editorial assistance in preparing the manuscript.
4Abbreviations used are as follows: CCNU, 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea; CR, complete response; LOH, loss of heterozygosity; PCV, procarbazine, CCNU (lomustine), and vincristine; PR, partial response.