Dose Reductions of Vincristine in Children with Medulloblastoma Treated in the Maintenance Arm of the Prospective Multicenter Trial HIT’91

Vincristine is known to have neurotoxic side effects (Sul JK and DeAngelis LM, Semin Oncol 2006; 33 (3): 324–332). Peripheral neuropathy is an important adverse reaction of vinca-alkaloids. Typically, it reflects an axonal peripheral neuropathy, but may involve the autonomic nervous system leading to abdominal pain, constipation, paralytic ileus, and sensory and motor dysfunction. Early symptoms include numbness and tingling of hands and feet accompanied by loss of deep tendon reflexes. The temporal relationship to treatment is variable (Sul JK and DeAngelis LM, Semin Oncol 2006; 33 (3): 324–332). The neurotoxicity of vincristine can be reversible on interruption of the therapy. Recovery is slow and takes several months (Legha SS, Med Toxicol 1986; 1 (6): 421–427). Although several compounds have been proposed as neuroprotective agents, few have been shown to be active against the chemotherapy induced neurotoxicity. Here, we aim to investigate whether vincristine leads to higher rates of toxicity-related dose reductions with increasing age of the children and adolescents.

Between August 1991 and December 1997, 280 patients 3–18 years of age with newly diagnosed medulloblastoma as well as children with other histologies were treated according to the prospective multicenter trial HIT’91 of the Society of Pediatric Oncology and Hematology of German-speaking countries GPOH (Kortmann RD et al., Int J Radiat Oncol Biol Phys 2000; 46 (2): 269–279; von Hoff et al., Eur J Cancer 2009; 45 (7): 1209–1217). Patients were randomly assigned to receive either immediate radiotherapy followed by ‘maintenance’chemotherapy, or immediate preradiation ‘sandwich’ chemotherapy starting within 14 days after surgery. Vincristine was given weekly concomitantly with radiotherapy (6–8 single doses of 1.5 mg/m2). Chemotherapy of patients in the maintenance arm corresponded to the ‘Packer’ protocol (Packer RJ et al., J Neurosurg 1991; 74: 433–440). The clinical rational for adopting this treatment schedule in HIT’91 was based on the favorable survival rates reported in a single institution trial (Packer RJ et al., J Neurosurg 1991; 74: 433–440), and excellent survival rates confirmed by expanding this treatment approach to three institutions (Packer RJ et al., J Neurosurg 1994; 81: 690–698). It was started six weeks after completion of radiotherapy and consisted of eight cycles with CCNU, vincristine (maximal 24 single doses of 1.5 mg/m2) and cisplatin. In case of distal paresthesias, motor weakness or colicky abdominal discomfort, the recommendation was to reduce vincristine dosage by 50%. Vincristine was omitted for at least one dose for ileus or CNS toxicities such as seizures. Criteria for modifications of cisplatin and CCNU (nephro-, oto-, hematotoxicities, infections) dosages were also defined.

A calculation of the cumulative doses per patient and according to age was not assessable, because the applied doses were documented in HIT’91 only for each application as follows: 100%, more than 75%, or less than 75% of the recommended dose applied. Maintenance chemotherapy was administered to 62 fully assessable patients with M0 or M1 disease. Six of 62 patients received less than four cycles for other reasons (progressive disease after the second cycle: 1 patient, parent's decision: 5 patients). These patients were excluded from the following analysis. The median age at diagnosis of the remaining 56 individuals was 10.3 years. As shown in [Fig. 1], in almost half of the patients there was a need for vincristine dose reduction (26 patients; median age 11.6 years). Twenty-one patients experienced neurotoxicity, whereas in some patients vincristine was reduced for other reasons. In contrast, the patients without dose reduction of vincristine were significantly younger (30 patients; median age 8.4 years; Mann-Whitney U test p=0.009). Overall survival of patients with (10-year OS 88±5%) and without (10-year OS 80±11%) any dose reduction of ‘maintenance’chemotherapy (CCNU, vincristine, and cisplatin) due to toxicity was not different (p=0.585, von Hoff et al., Eur J Cancer 2009; 45 (7): 1209–1217). In summary, our data suggest that there may be an association of vincristine-tolerability and age.

Fig. 1 Non-parametric T test comparing the age of patients with (filled square: 26 patients; median, 11.6 years) and without (open square: 30 patients; median, 8.4 years) vincristine dose reductions (reduced number of absolute vincristine applications). Patients with vincristine doses reductions were significantly older than patients without vincristine doses reductions (p=0.009) as determined by the Mann-Whitney test.

Zeltzer et al. treated in one arm of their phase III (CCG 921) randomized study 96 high-stage medulloblastoma patients using a similar treatment regimen (Zeltzer et al., J Clin Oncol 1999; 17 (3): 832–845): Postoperative therapy consisted of radiotherapy (XRT, 54 Gy tumor/36 Gy neuraxis) with eight weekly applications of vincristine (1.5 mg/m2 injection during XRT) followed by a ‘maintenance’ chemotherapy with eight 6-week cycles of VCR 1.5 mg/m2 weekly for 3 weeks, lomustine 100 mg/m2 for 1 day, and prednisone 40 mg/m2 for 14 days. During induction therapy, grade 3 or 4 peripheral neurotoxicity was observed in 7 of 96 patients. However, after ‘maintenance’ chemotherapy in 46/96 patients one or more episodes of grade 3 or 4 peripheral neurotoxicity were reported. Thus, using a similar treatment regimen in terms of vincristine application, vincristine dose reductions were as frequent as in our study. However, no information about the age of patients was given. Greenberg et al. reported that the toxicity of the ‘Packer’ protocol in adults with medulloblastoma (10 patients) appears to be much greater. In 90% of adult patients there was a need for vincristine dose reduction, 60% of the patients received less than 3 cycles of vincristine (Greenberg HS et al., Neuro Oncol 2001; 3 (1): 29–34). Another study showed that 27 of 40 patients (age between 20 and 75 years) with Non-Hodgkin's Lymphoma treated with vincristine (1.4 mg/m2; cumulative dose 12 mg in 18–24 weeks) presented with signs and symptoms of vincristine neuropathy (Postma TJ et al., J Neurooncol 1993; 15 (1): 23–27), providing further evidence that age may play a role in the vincristine-tolerability.