Interleukin-8 is not a predictive biomarker for the development of the acute promyelocytic leukemia differentiation syndrome

Differentiation syndrome (DS) is a life-threatening adverse event that occurs in approximately 20–25% of patients with acute promyelocytic leukemia (APL) undergoing induction therapy with all-trans retinoic acid (ATRA) [1, 2]. During the APL-DS, changes in seric levels of cytokines [3], and in cellular adhesion/migration properties [1, 4, 5], as well as endothelial damage [6] have been reported to be related to ATRA-driven DS. The in vitro release of pro-inflammatory cytokines such as interleukin (IL)-1β, IL-6, IL-8, and tumor necrosis factor-α (TNF-α) have been reported to coincide with ATRA-induced differentiation of APL blasts, which may lead in vivo to a systemic inflammatory response syndrome (SIRS) [3, 7]. Moreover, the ATRA-differentiating APL cells have increased ability to migrate from the blood flow into the tissues by upregulating molecules involved in cell adhesion (e.g.: ICAM: Intercellular Adhesion Molecule-1 and 2) and migration (e.g.: E-selectin and β2-integrin) [5, 8].

The lung is the most clinically relevant target organ in APL-DS patients, which can develop distinct pulmonary complications [9]. Accordingly, it was experimentally demonstrated that increased expression of ICAM-1 on the lung of NOD/scid mice after ATRA therapy is important for the pulmonary infiltration of APL cells [10]. Moreover, the IL-8 secretion by A549 alveolar epithelial cells support the chemotactic transmigration of ATRA-treated NB4 APL cells toward A549 cells [11]. In fact, IL-8 plays a important role in acute inflammation by activating and chemoattracting neutrophils [12]. Patients with acute myeloid leukemia (AML) that express lower levels of IL-8 have better survival outcomes [13]. Accordingly, the IL-8 receptor CXCR2 is an adverse prognostic factor in AML and its inhibition decreases the proliferation of AML cell lines and primary samples [14].

Despite the lack of biomarkers that predict development of DS, Shibakura et al. observed that serum levels of IL-8 were increased during the course of ATRA treatment in two patients who developed APL-DS [15]. To support these in vivo findings, the same authors confirmed that the in vitro IL-8 expression was also up-regulated in leukemic primary cells from both patients with APL after incubation with ATRA [15]. Nevertheless, no previous study has investigated and compared the IL-8 plasma levels in the plasma of DS and non-DS patients with APL during the treatment with ATRA to determine whether this cytokine has the potential to predict the development of the APL-DS.

Seventeen patients with APL (10 females, 7 males) with an age range of 19–72 years (median age, 36 ± 16 years) diagnosed at the Clinics Hospital of Ribeirão Preto (HCFMRP), University of São Paulo, from March 2007 through July 2013 were included in this study. All the six patients with DS presented with dyspnea, pulmonary infiltrates and unexplained fever and, in two cases there was evidence of kidney failure as well (severe DS). The symptoms started between 9th and 18th day of ATRA therapy. All but one recovered from the DS and achieved and remain in complete remission. Eight healthy controls (8 females, 3 males) were included as controls. Samples were collected only from control subjects with no history of fever within 1 week, use of any medications or drugs, pregnancy, and chronic diseases. In our cohort, IL-1β, IL-10, IL-12p70 and TNF-α were not detected in the plasma of patients with APL regardless of the development of DS. These findings may due to the absence of such cytokines in the samples or to the fact that these cytokine concentrations are below the detection limit of the CBA (IL-1β: < 7.2 pg/mL; IL-10: < 3.3 pg/mL; IL-12p70: < 1.9 pg/mL; TNF-α: < 3.7 pg/mL).

Table 1 shows the levels of IL-6 and -8 at D0, after three and 7 days of treatment. IL-6 and IL-8 were not detected in healthy control group (data not shown). The distribution of the plasma concentration levels of IL-6 and IL-8 at D0 ranged from 0 to 83.74 pg mL^− 1 and from 0.83 to 238.60 pg mL^− 1, respectively. Although the median values of IL-6 and IL-8 were lower in the group of patients who developed DS compared to the non-DS group the differences were not significant (Fig. 1a-b; Table 1; Supplementary file 1; both p > 0.05). At the D3, the median value of IL-6 concentrations was higher in the DS group, but again the difference was not significant (Fig. 1a-b; Table 1; Supplementary file 1; p = 0.60). In contrast, at the seventh day of treatment with ATRA, we observed that APL plasma samples in non-DS group exhibited significant decreased IL-8 levels (34.16; 6.99 to 147.11 pg mL^− 1 in D0 vs. 10.9; 0 to 26.81 pg mL^− 1 in D7; p = 0.02 - Fig. 1b; Table 1; Supplementary file 1). Although there was a decrease in IL-8 levels in DS-group it was not significant (23.09; 2.63 to 79.64 pg mL^− 1 in D0 vs. 6.2; 0 to 17.24 pg mL^− 1 in D7; p = 0.06 - Fig. 1b; Table 1; Supplementary file 1). In addition, there were no differences between the two groups in the IL-6 levels throughout treatment (Fig. 1a; Table 1; Supplementary file 1; all p > 0.05). Of note, assessing our cohort, only 6/17 (2 from DS- and 4 from non-DS group) and 2/17 (1 from DS- and 1 from non-DS group; Supplementary file 1) patients with APL, respectively, for the levels of IL-6 and IL-8 became undetectable at D7, such as observed in healthy controls. Moreover, our study not detected significant differences between IL-6 and IL-8 levels at D0/D3 (Fig. 1c; Supplementary file 1; both p > 0.05) and D0/D7 (Fig. 1d; Supplementary file 1; both p > 0.05) ratios when comparing DS- and non-DS groups.

Table 2 shows the distribution of the main clinical and laboratory variables in the groups of patients with low or high levels of IL-6 and IL-8 at diagnosis. Patients with WBC counts higher than 10⁹/L presented lower concentrations of IL-8 at diagnosis. In fact, with the exception of WBC counts (p = 0.04; Table 3; Supplementary file 1), the clinical features and laboratory results of patients with APL at D0 were not associated with DS development (all p > 0.05; Table 3; Supplementary file 1). Patients who develop (median value 6.2 ± 5.9 pg mL^− 1) or not (median value 10.9 ± 7.9 pg mL^− 1) the DS persisted with detectable low plasma concentrations of IL-8 after 7 days of treatment with ATRA. Finally, the variation of IL-8 levels in patients with APL following ATRA treatment occurred independently of the development or not of DS (p = 0.002; D0 vs. D7 of DS- and non-DS groups combined together - Fig. 1b; Supplementary file 1).

The absence of IL-1β, IL-10, IL-12p70 and TNF-α expression in our cohort of patients with APL, observed in the groups with and without DS, highlights the difference in cytokine profiles of APL and non-APL AML because Turzanski et al. and Sanchez-Correa et al. have reported that IL-1β and IL-10 are detected in the plasma of patients with AML without t(15;17) and may play a role in apoptosis-resistant phenotype and clinical outcome, respectively [18, 19].

In this study, among the cytokines examined, IL-6 and IL-8 were the only ones detected on the plasma of patients with APL, but differentially modulated by treatment with ATRA over time. Similar to our results, Dubois et al. [1] demonstrated that the incubation of APL primary cells with ATRA did not modulate the IL-6 production but significantly decrease the levels of IL-8 in the supernatant. In addition, in contrast to the previous report, showing an relationship between high IL-8 serum levels and the occurrence of DS in two patients with APL [15], our data demonstrate that patients who develop or not the DS may persist with detectable low plasma concentrations of IL-8 after the first week of treatment, suggesting that the modulation of IL-8 levels in patients with APL following ATRA treatment may occur regardless of the DS development.

A key problem in the diagnostic process of DS is the lack of precise definitions of clinical criteria and biomarkers. In addition, the DS diagnosis is often challenging when the signs and symptoms attributable to DS occur in patients with APL who also develop complications such as pneumonia, cardiac toxicity, renal failure, and septic shock [6]. Considering the APL-DS patients with apparently no other complications, Montesinos et al. proposed criteria for APL-DS severity grading based on the presence of predefined signs and symptoms [20] but this classification has not been widely adopted in practice as occurred in the current study once the number of patients enrolled in the cohort did not allow a stratified analysis of the DS-group. Although the lower serum levels of IL-8 has been reported to be associated with longer survival in AML patients [13], the prognostic relevance of IL-8 in APL remains to be established in larger cohorts that accurately stratify the patients with APL for severity ranking of the DS.

The present study demonstrated that ATRA treatment reduces the levels of IL-8 regardless of the occurrence of DS and, therefore, our findings do not support that IL-8 is a predictive biomarker for monitoring the development of the APL-DS.