Background: In what to the authors' knowledge is the first such study for a pediatric cancer, a large database was retrospectively analyzed to assess statistically the likelihood of response to a given salvage therapy in different clinical subsets of patients.
Methods: Treatment was comprised of high-dose cyclophosphamide (at a dose of 140 mg/kg), topotecan (at a dose of 8 mg/m(2)), and vincristine (at a dose of 0.067 mg/kg or 2 mg/m(2), whichever was lower; maximum dose, 2 mg) (HD-CTV). The Fisher exact test was used for comparisons of response rates among standard subsets of patients (n = 126) with refractory or recurrent neuroblastoma (NB).
Results: Among children, major (ie, complete/partial) responses occurred in 11 of 58 (19%) with primary refractory NB, 4 of 14 (29%) with secondary refractory NB, 13 of 25 (52%) with a new (first) disease recurrence, and none of 13 patients with progressive disease (PD) while receiving therapy. Other children had mixed responses (MRs); when combining major responses and MRs, anti-NB activity was noted in 26 of 58 (45%) children with primary refractory NB, 10 of 14 (71%) children with secondary refractory NB, 20 of 25 (80%) children with a new (first) disease recurrence, and 1 of 13 (8%) children with PD while receiving therapy. The response rate was significantly different across the 4 groups of children for both major responses (P = .003) and combined responses (P = .001). All 10 adolescents/adults treated for primary refractory NB had no response, which was a significantly inferior result compared with the response rate of 45% noted in children with primary refractory NB (P = .008).
Conclusions: Response to HD-CTV as salvage therapy is significantly less likely in adolescents/adults and in children with NB that is persistent or progressing on treatment rather than newly recurrent off treatment. These findings are broadly applicable and should be considered when designing, and interpreting the results of, phase 2 studies.