Abstract
Background: Currently, pediatric AML risk stratification relies solely on response to therapy and a small number of prognostic genetic factors. With these limited criteria, approximately 75% of children are initially classified as being at a lower risk (LR) of relapse. However, 30% of LR patients relapse, and achieving a second remission is often quite difficult. Therefore, there is an urgent need to understand mechanisms of chemoresistance, in order to identify patients who should receive alternative therapies. We previously showed that increases in IL-6-induced STAT3 activity at relapse were associated with inferior survival [Stevens, et al, Haematologica, 2015]. These findings suggested that the IL-6-induced STAT3 pathway may promote the chemoresistance often seen in relapsed AML. Thus, we hypothesized that IL-6 levels in the bone marrow (BM) niche are dysregulated in pediatric AML and elevated levels are predictive of clinical outcome. We further hypothesized that IL-6 promotes chemoresistance in AML blasts in vitro.
Methods: Bone marrow samples from pediatric AML patients diagnosed between 2007-2016 at Texas Children's Cancer Center were studied. Patients were treated similarly (Children's Oncology Group protocols AAML0531 or AAML1031). We obtained plasma from diagnostic BM for 46 patients, and 23/46 had a matched remission sample. For comparison, we obtained normal BM (NBM) plasma from 7 sibling BM donors. Cytokine levels were determined with the 41-plex Milliplex MAP Human Cytokine/Chemokine Magnetic Bead Panel (EMD Millipore). Cytokine and growth factor concentrations between NBM and AML samples were analyzed by the Mann-Whitney U test. Paired diagnostic and remission samples were compared with the Wilcoxon signed-rank test, and correlation of cytokine levels with clinical outcome was assessed by the Kaplan-Meier method. To assess for IL-6-induced chemoresistance, AML cell lines NB4 and THP1 were used. Cells were exposed to IL-6 (50 ng/mL) and soluble IL6 receptor a (sIL-6Rα; 100 ng/mL) or vehicle control (VC) for 24h prior to escalating doses of chemotherapy (mitoxantrone, etoposide, cytarabine). After treatment with chemotherapy for 24h, the percent of Annexin-V+ AML cells was determined by FACS and differences assessed by ANOVA.
Results: IL-6 levels were uniformly low in NBM and were elevated at diagnosis in a subset of AML patients. The median in NBM was 1.7pg/mL (25-75 Percentile 1.53-4.54) vs 11.84 pg/mL for AML (25-75 Percentile 4.66-41.47, p<0.001). Fevers at diagnosis were frequent 30/46 and correlated weakly with IL-6 levels (Spearman R=0.33, p=0.03), but did not predict outcome. Infection at the time of diagnosis was rare and not associated with elevated IL-6 (n=6). IL-6 levels were reduced in paired remission samples (median 2.8 pg/mL, 25-75 Percentile 1.7-3.1) vs diagnostic samples (median 11.79 pg/mL, 25-75 Percentile 4.78-48.0, p=0.0015). To determine whether elevated IL-6 levels at diagnosis were associated with an increased risk of relapse, we analyzed outcomes for patients who did not receive stem cell transplant in their first remission (n=29). Here, cut-point analysis revealed increased relapse rates in patients with BM IL-6 levels > 17.5 pg/mL. Patients with high IL-6 (n=11) had relapse free survival (RFS) of 13.6% compared to 61.5% (p=0.007; log-rank test) for patients with low IL-6 (n=18). In addition, we found the following cytokines and growth factors to be significantly higher in AML patients v. controls: IL-8, IL-10, GM-CSF, IL-3, IL-17A, MDC, PDGF-AA, PDGF-BB, IL-1RA, IL-4, and TNFα. Notably, and discrepant from what is seen in adult AML, only IL-6 was found to predict RFS in our cohort. Furthermore, both THP-1 and NB4 cells demonstrated less mitoxantrone-induced apoptosis when pretreated with IL-6 and sIL6Rα, compared to cells treated with VC.
Conclusions: Our data demonstrate that in pediatric AML, IL-6 levels in the BM at diagnosis are elevated in a subset of patients and return towards normal levels in the majority of patients at remission. In our cohort, elevated IL-6 levels in the BM at diagnosis identified patients with an inferior outcome. Additionally, we have demonstrated that these inferior outcomes may be due to IL-6-induced chemoresistance. Our results provide support for further development and evaluation of IL-6 as a marker of prognosis and as a potential therapeutic target in pediatric AML.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal