To the editor:
We read with interest the paper by Yanik and colleagues describing the impact of using soluble tumor necrosis factor (TNF)–α binding protein, etanercept (Enbrel; Amgen, Thousand Oaks, CA) on the treatment of idiopathic pneumonia syndrome (IPS) following allogeneic hematopoietic stem cell transplantation (HSCT).1 The authors describe their experience with 15 patients who developed IPS at a median time of 14 days after HSCT. The overall survival at days 28 and 56 from the first etanercept dose is reported to be 73% and 60%, respectively. One of the main conclusions of the paper is that the combination of corticosteroids and etancercept improves survival. We believe that the data provided support improvement in early survival but did not contribute to improving long-term survival.
Using the data provided in Table 1 of the article by Yanik et al,1 we used the Kaplan-Meier method to estimate the overall survival at 100 days and at 1 year after diagnosis of IPS for the patients described. The confidence intervals (CIs) are based on the log hazard. The estimated survival at 100 days and at 1 year is 47% (95% CI: 21, 69) and 20% (95% CI: 5, 42), respectively. Twelve of the 13 deaths reported were related transplantation-related causes and 1 was related to relapsed disease. The same group reported on 3 pediatric patients treated with etanercept for IPS, and all 3 patients responded to the initial therapy but died before 120 days after transplantation (2 related to organ dysfunction and 1 related to relapsed disease).2
We have reported the outcome with 11 pediatric patients who developed IPS at a median of 17 days after allogeneic HSCT.3 Five of the 11 patients required assisted ventilation. All patients were treated with corticosteroid, and those who did not respond to corticosteroid therapy were treated with etanercept (n = 3) or inflixamab (Remicade; Centocor, Malvern, PA; n = 3). Nine of the 11 patients (81%) had resolution of respiratory symptoms and were weaned off oxygen. The overall survival at 100 days and at 1 year after the diagnosis of IPS was 73% (95% CI: 37, 90) and 32% (95% CI: 8, 60) respectively. A large study by the Seattle group reported on the outcome of 81 patients with IPS among 1100 patients who underwent myeloablative or nonmyeloablative HSCT.4 Despite aggressive supportive care, the 100-day and 1-year survival after diagnosis of IPS was 23% and 17%, respectively. None of the patients were treated with anti-TNF therapy.
Although treatment with corticosteroids and etanercept improved the short-term survival (day 56) for patients with IPS, the overall survival at 1 year was not impacted by this therapy compared with patients who only received supportive care. The development of lung injury after allogeneic HSCT represents a poor prognostic indicator regardless of the response to the initial therapy. Future strategies should not only address the early mortality associated with IPS but also the long-term high rate of transplantation-related mortality in this subset of patients.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Haydar Frangoul, MD, Vanderbilt University, 220 Pierce Ave, 397 PRB, Nashville, TN 37232-2573; e-mail: Haydar.Frangoul@Vanderbilt.edu.