4Sc-205, First Oral KIF11 Inhibitor, Exhibits Antitumor Activity and Potentiates Standard and Targeted-Therapies in Primary and Metastatic Neuroblastoma Models

Dear Editor, Neuroblastoma remains incurable for most patients with high-risk disease.1 Perturbation of transcription factors (MYCN and PHOX2B), kinases (ALK, MEK), and cell cycle regulators (CDK4/6, CHECK1), among other factors, make neuroblastoma cells highly proliferative, which is associated with poor patient outcomes.2, 3 To circumvent the limitations of the classical microtubule poisons such as vinca alcaloyds used in the treatment of neuroblastoma,1 we sought to explore alternative mitotic regulators as new therapeutic targets for high-risk neuroblastoma patients. One of these mitotic spindle-specific proteins is kinesin family member 11 (KIF11), also known as kinesin spindle protein, kinesin-5, or Eg5, which is essential for bipolar spindle formation and mitotic progression in human cells.4 Transcriptomic analyses showed that the expression of multiple kinesins, including KIF11 was higher in the high-risk neuroblastoma compared with low- and intermediate-risk groups (Figures 1A and S1; Table S1). Overall survival was significantly poorer in patients with high KIF11 expression (Figure 1B–D; Table S2). KIF11 high expression was identified as an independent prognostic factor of survival, together with risk assessment (HR = 3.051; Table S3) and found to be higher in patients with amplification of MYCN, 1p36 loss, or 17q23 gain (Figure 1E). At the protein level, KIF11 expression was detected in the cytoplasm of neuroblastic cells (Figure 1F) and showed higher expression compared to low- or intermediate-risk neuroblastoma samples (p < 0.05) and in tumors with segmental chromosome alterations such as 1p36, 11q deletion, and gain of 17q23 (Table S4). Kaplan–Meier analysis confirmed that high KIF11 protein expression was associated with shorter event-free and overall survival (Figures 1G and H). While there is a positive correlation between KIF11 and MYCN mRNA expression levels, MYCN is neither sufficient, nor necessary for KIF11 expression (Figure S2). According to functional genomics, neuroblastoma cells seem to be one of the cell types that are more dependent on the expression of KIF11 for survival being particularly sensitive to its pharmacological inhibition.5 Concurring with these observations, the silencing of KIF11 caused a reduction in cell viability (Figure S3A-C) and a 3–4 fold reduction in the growth of established neuroblastoma subcutaneous xenografts (Figure 2A–C and S3D-R). KIF11 inhibitors have moved forward toward phase 1 and 2 clinical trials in adult tumors,6, 7 with very limited development for childhood cancer. Herein, we provide a complete preclinical characterization of the potent and highly selective KIF11 inhibitor, 4SC-205 (Figure 2D), the first oral KIF11 inhibitor that has been evaluated in phase I clinical trials in adult patients (NCT01065025). Compared to other KIF11 inhibitors, 4SC-205 can be administrated daily, thus being able to hit the target in a more sustained manner. Neuroblastoma cells treated with 4SC-205 (Figure 2E; Table S5) displayed all the expected phenotypic features resulting from KIF11 inhibition such as the inability to form bipolar spindles (Figures 2F and S4A), cell cycle arrest during mitosis (Figure S4B-H), and induction of apoptosis (Figure S5), thereby confirming the high KIF11 specificity of this compound. While similar effects were observed in 3D spheroid cultures (Figure S6A-C), 4SC-205 did not affect the viability of differentiated cells (Figure S6D-G). When used in vivo, 4SC-205 treated mice showed a remarkable shrinkage of the original SK-N-BE(2) subcutaneous xenograft (Figures 2G and 2H) or tumor growth delay in SK-N-AS xenografts (Figures 2J, 2K, and Figure S7). Increased phosphorylation of histone H3 and apoptotic hallmarks (i.e., processing of PARP) confirmed that the antitumor effect of 4SC-205 was comparable to that of genetic KIF11 silencing in vivo (Figures 2I,L). Transcriptomic analysis of the 4SC-205-treated tumors confirmed the expected genetic changes of arrested cells in mitosis and tumor cells with reduced proliferation or viability (Figure 2M–O). We next tested 4SC-205 in a patient-derived orthotopic xenografts (PDOX) derived from a very high-risk neuroblastoma patient. VH-NB608 PDOX retained most of the histological and molecular features of the original tumor (Figure 3A–D). 4SC-205-treated mice displayed a 14.75-fold reduction in tumor weight compared to the vehicle group (Figures 3E and 3F). Mice treated with 4SC-205 presented small tumors located in the adrenal gland, whereas vehicle-treated mice had large tumors with the kidney completely surrounded by the tumor (Figure 3G). Furthermore, 4SC-205 tumors had a larger fraction of cells with phosphorylation of histone H3, thereby indicating a specific targeting of KIF11 in these tumors, and suggesting that tumors were still sensitive to the inhibitor after 3 weeks (Figure 3H,I). Half of neuroblastoma patients present metastases at the time of diagnosis.8 Therefore, we proceeded to test the efficacy of 4SC-205 in a neuroblastoma liver metastasis model. In response to treatment, a clear delay in metastatic outgrowth was observed in 4SC-205-treated mice (Figure 3J–L and S8). As a consequence, the median lifespan of the animals was significantly expanded by ∼27% (Figure 3M; vehicle: 33 days vs. 4SC-205: 42 days). Noticeable, 4SC-205 administration minimally affected mice weight (<10%) during the course of the treatment (Figure S9). To achieve a better therapeutic effect and provide a rationale for further development of 4SC-205 in clinical trials, we combined 4SC-205 with chemotherapies, such as platine derivatives (cisplatin), doxorubicin, and topotecan, which are currently used as standard treatment for patients with high-risk neuroblastoma. In all cases, the combination of 4SC-205 with the chemotherapies showed additive effects (Figures 4A–D and S10A-C; Table S6). Pediatric precision medicine programs have discovered a small number of recurrent alterations such as ALK activating mutations or hyperactivation of the ERK Pathway,9, 10 which constitute the basis for the development of targeted therapies against high-risk neuroblastoma tumors. Thus, we combined 4SC-205 with two ALK inhibitors (ceritinib and lorlatinib) or with the MEK1/2 inhibitor selumetinib. The combination of 4SC-205 with ALK or MEK inhibitors showed a ∼2–3-fold reduction in cell proliferation compared with the inhibitors alone, with most of the combination doses showing additive effects (Figures 4E–J and S10D-G; Table S7). In summary, our study provides a rationale for the future therapeutic integration in clinical trials of 4SC-205, an structurally distinct oral KIF11 inhibitor that shows potent antitumor activity in multiple preclinical neuroblastoma models and sensitizes neuroblastoma cells to standard chemotherapy and specific neuroblastoma-targeted therapies. We are very thankful to Prof. Thomas U. Mayer for providing us with contact with the 4SC. We would like to acknowledge the members of the Laboratory Animal Service and High Technology Unit for technical support. We are grateful to the CNAG-CRG for technical and bioinformatics assistance with transcriptome analyses. We would like to thank Editage (www.editage.com) for English language editing. Dr. Moreno participates in data monitoring committees of clinical trials sponsored by Novartis, Actuate Therapeutics, Shionogi, Incyte, the University of Southampton and the Royal Marsden NHS Foundation Trust; and had a consulting role for Novartis and Shionogi. Dr. Lucas Moreno is also a member of the Executive Committee of the European neuroblastoma research cooperative group (SIOPEN) which receives royalties for the sales of dinutuximab beta. Rolf Krauss (RK) is an employee of 4SC. Alberto Villanueva (AV) is co-founder of Xenopat S.L. No potential conflict of interest was disclosed by the rest of the authors. Miguel F. Segura and Anna Santamaria conceived and designed the study. Marc Masanas, Nuria Masiá, Leticia Suárez-Cabrera, Mireia Olivan, Aroa Soriano, Blanca Majem, Carlos Jimenez, Ariadna Boloix, Ignasi Toloedano, Gabriela Guillén, Alexandra Navarro, and Alberto Villanueva carried out the experiments. Marc Masanas, Laura Devis-Jauregui, Rebeca Burgos-Panadero, Pau Rodriguez-Sodupe, and Aroa soriano analyzed the data. David Llobet-Navas, Josep Sánchez de Toledo, Josep Roma, Rosa Noguera, Lucas Moreno, and Soledad Gallego provided intellectual support for result interpretation and critical revision. Rolf Krauss provided the compound used in this study. Marc Masanas, Miguel F. Segura, and Anna Santamaria wrote the initial manuscript. All the authors read and approved the final manuscript. The authors confirm that all data supporting the findings of this study are available within the article and the corresponding web servers. Further information from WES data analyses is available from the corresponding authors upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.