Unsere Arbeitsgruppe wird seit April 2011 von der Deutschen Krebshilfe e. V. im Rahmen des Max-Eder-Nachwuchsgruppenprogramms neu gefördert.
Our long-term goal is to elucidate the mechanisms driving dormancy during minimal residual disease and causing disease recurrence in pediatric ALL. Despite good cure and overall survival rates, roughly 15% of pediatric ALL patients eventually develop relapse and some of these relapses may occur years after termination of therapy or, at times when residual disease was not detectable by routine diagnostic methods shortly before. This suggests that dormant leukemia cells survive in growth-restrictive niches, and, upon yet unknown cues, resume proliferation thus causing clinically detectable medullary and/or extramedullary disease. Late relapses generally occur in patients with good initial treatment response and often in subgroups with favorable genetic alterations such as t(12;21) or t(1;19). Early relapses tend to develop in other genetically defined subgroups such as patients with MLL-rearrangement (e. g. t(4;11) or t(9;11)) and BCR-ABL1 fusion (t(9;22)). The fact that ALL relapse remains the fourth most common cancer in children and a leading cause of cancer-related mortality, emphasizes the need for new therapeutic strategies further improving and optimizing long-term outcomes in defined subsets of patients. Patients harboring dormant leukemia may be at high risk for relapses and current risk group stratification by initial therapy response may not be sufficient for their detection.
On the cellular level, dormancy is defined by a reversible state of growth arrest and survival. Previous work by our group has shown that tumor cell dormancy in head and neck cancer (HNSCC) can occur through activation of the p38, TGF-beta and endoplasmic reticulum (ER) stress pathways. These signals maintain cell cycle arrest and promote survival upon chemotherapy and microenvironmental stress, e. g. upon dissemination. Recent work coauthored by our group suggests that the degree of permissivity of the microenvironment largely regulates the fate of disseminated tumor cells (DTCs): In “restrictive” sites such as the bone marrow, high TGFβ2 signaling intensity induces a p38high dormant phenotype while “permissive” sites such as the lung display weak TGFβ2 signals and a p38low proliferative phenotype in human HNSCC xenografts. The situation in ALL may be similar, as ALL cells are highly influenced by niche-dependent growth signals. This is supported by multiple experiences that reliable in vitro culture of most ALL primary samples is virtually impossible. However, as ALL is bone-marrow derived, microenvironments “restrictive” to DTC cell growth may be “permissive” to ALL and vice versa. It has for instance been demonstrated that the production of important pro-proliferative cytokines in co-cultures of ALL cells with bone marrow mesenchymal stromal cells (BMSCs) such as CXCL12, IL-6, VEGF and PDGF is p38 dependent. However, these stromal cells are immortalized and may have gained cancer-like properties so that observations from co-culture models may not reflect the reality in vivo adequately.