Disease-Associated Changes In The Repair Efficiency Of Double Strand Breaks Affect Melphalan Sensitivity Of The Bone Marrow Plasma Cells and Correlate With The Clinical Outcome Of Anti-Myeloma Therapy

Melphalan is an interstrand cross-link (ICL)-inducing agent and one of the most active chemotherapeutic drugs in the treatment of patients with multiple myeloma (MM). There is clear evidence that the formation and subsequent persistence of ICL correlates with its cytotoxicity. Previous studies have established that during ICL repair, replication forks stall at the ICL inducing the formation of a lethal form of DNA damage (DNA double-strand breaks, DSBs), which is repaired mainly by homologous recombination (HR) and non-homologous end joining (NHEJ). In this report, we investigated the molecular mechanisms of therapeutic efficiency and drug resistance to ICL-inducing agents using melphalan as a model.

We studied two MM cell lines (melphalan-sensitive RPMI-8226 and melphalan-resistant RPMI-LR5) and 70 MM patients (38 males/32 females; median age 59 years) who underwent high-dose melphalan (HDM) therapy with autologous stem cells transplantation (ASCT) as first line therapy. Patient response status was assessed 100 days after ASCT according to the International Myeloma Working Group Criteria; patients were grouped into responders (≥PR, n=48) and non-responders (<PR, n=22). Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples obtained from MM patients, at diagnosis or at least 1 week prior to the treatment with any anti-myeloma drug. In addition, bone marrow plasma cells (BMPCs) were isolated from bone marrow trephine aspiration samples during diagnostic clinical assessment. Primary cells (PBMCs and BMPCs) and MM cell lines were ex vivo treated with melphalan either alone or in combination with RI-1 (selective inhibitor of HR) or NU7026 (selective inhibitor of NHEJ) and the extent of the N-ras-specific ICLs and DSBs (intermediates of ICL repair) were evaluated using a quantitative PCR assay and quantification of γH2AX foci, respectively. The γH2AX foci were viewed under a laser-scanning confocal immunofluorescence microscope and quantitated using Image J software. The induction of the apoptotic pathway by melphalan, using a photometric enzyme-immunoassay, was also studied.

Following ex vivo treatment of BMPCs with melphalan, ICLs reached maximal levels within 8h of the melphalan treatment. Thereafter, ICLs levels were reduced with the repair efficiency being significantly higher in non- responders (half-time of damage removal, t_1/2 23h) than in responders (t_1/2 48h) (P<0.01). Moreover, γ-H2AX foci formation followed the timing of ICL formation and reached maximal levels within 8h. Thereafter, γ-H2AX foci levels declined rapidly, suggesting the resolution of the intermediate DSBs by downstream pathways (HR, NHEJ). Interestingly, the repair efficiency of DSBs in BMPCs was significantly higher in non-responders (t_1/2 9h) than in responders (t_1/2 12h) (P<0.02). Similar results were obtained using PBMCs.

Also, in both BMPCs and PBMCs, the melphalan-induced apoptosis inversely correlated with the repair efficiencies of ICLs and DSBs, with the toxicity being higher in responders than in non-responders (P<0.01). Moreover, RPMI-LR5 cells showed higher repair efficiencies of both ICLs and DSBs and lower toxicity than RPMI-8226 cells. Interestingly, in all cell types analyzed, significant correlation between ICL and DSBs levels was observed (linear regression analysis, R²=0.67, P<0.01).

To further elucidate the mechanism of drug-induced DSBs repair, MM cell lines and primary cells (BMPCs and PBMCs) were treated with melphalan in combination with nontoxic doses of RI-1 or NU7026. We found that the combined treatment of melphalan with RI-1 or NU7026 significantly increased the melphalan only-induced phosphorylation of H2AX (suggesting that both HR and NHEJ contribute to the repair of melphalan-induced DSBs), delayed the repair of ICLs and strongly enhanced the cytotoxic activity of melphalan (all P<0.01).

Collectively, these results highlight that in BMPCs significant changes in the repair efficiency of DSBs occur in MM patients. These changes affect the removal of the cytotoxic ICLs, modify drug sensitivity of the malignant plasma cells ex vivo, and correlate with the clinical outcome of anti-myeloma therapy. Interestingly, these changes are also reflected in PBMCs. Specific inhibition of HR and/or NHEJ may be useful as an adjunct to melphalan therapy in MM patients.