The most important complication of all-trans retinoic acid (ATRA) treatment in acute promyelocytic leukemia (APL) is the retinoic acid syndrome (RAS), a life-threatening hyper-inflammatory reaction with a distinct capillary leak syndrome and multi-organ failure. Once established, the syndrome has proven very difficult to manage. Early recognition in combination with prompt corticosteroid treatment significantly reduces the mortality rate of patients with this syndrome. Nonetheless, still 15% of the induction deaths in APL is caused by RAS. The pathogenic mechanism of the hyper-inflammatory cascade in RAS is not fully understood yet. Infiltration of differentiating APL cells in the lung is important in the development of RAS. In addition, differentiation of APL cells by ATRA is associated with increased expression and release of pro-inflammatory cytokines, chemokines and adhesion molecules in vitro and it has been hypothesized that this is important for the triggering of the hyper-inflammatory cascade in RAS. We investigated the effect of ATRA on the expression and secretion of chemokines in APL cells in vitro and in vivo. Using microarray, quantitative PCR and ELISA we demonstrated significant induction (up to 16000-fold) of 8 CC-chemokines (CCLs) in the NB4 APL cell line after differentiation induction with ATRA in vitro. To demonstrate the significance of chemokine induction by ATRA in vivo, we measured plasma levels of CCLs in 3 APL patients, treated with a combination of ATRA, idarubicine and prednisone according to the AIDA-2000/P protocol. One of these patients developed an obvious retinoic acid syndrome. During therapy we observed increased plasma levels of 3 CCLs in all three patients. Induction of 5 other CCLs (CCL3, CCL4, CCL7, CCL8, CL11) was only observed during RAS, suggesting that upregulation of these 5 CC-chemokines is specific for RAS. To investigate whether the plasma levels of CC-chemokines are functionally relevant, we measured chemo-attraction of peripheral blood leukocytes towards plasma of an APL patient who developed RAS using a transwell system. Plasma from this patient during RAS showed significant more chemo-attraction than plasma from the same patient before treatment with ATRA, idarubicine and prednisone. Subsequently we investigated whether the therapeutic effect of dexamethasone in RAS can be ascribed to downregulation of chemokine expression in APL cells. Despite dexamethasone, the induction of chemokine expression in NB4 cells by ATRA sustained. We conclude that ATRA causes massive secretion of chemokines by APL cells, which might trigger the hyper-inflammatory cascade in RAS by continuous attraction of APL cells and other inflammatory cells towards tissues like the lung. Dexamethasone does not abrogate the production of CC-chemokines by APL cells, but rather seems to inhibit the hyper-inflammatory cascade at the level of the effector cells and target tissues like the lung. This might explain why dexamethasone is not able to sufficiently reverse a retinoic acid syndrome once it has been established. The application of neutralizing CC-chemokine receptor antibodies or other antagonists might be an alternative route to treat an established retinoic acid syndrome.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author