Drug response in a genetically engineered mouse model of multiple myeloma is predictive of clinical efficacy

Abstract

The attrition rate for anticancer drugs entering clinical trials is unacceptably high. For multiple myeloma (MM), we postulate that this is because of preclinical models that overemphasize the antiproliferative activity of drugs, and clinical trials performed in refractory end-stage patients. We validate the Vk*MYC transgenic mouse as a faithful model to predict single-agent drug activity in MM with a positive predictive value of 67% (4 of 6) for clinical activity, and a negative predictive value of 86% (6 of 7) for clinical inactivity. We identify 4 novel agents that should be prioritized for evaluation in clinical trials. Transplantation of Vk*MYC tumor cells into congenic mice selected for a more aggressive disease that models end-stage drug-resistant MM and responds only to combinations of drugs with single-agent activity in untreated Vk*MYC MM. We predict that combinations of standard agents, histone deacetylase inhibitors, bromodomain inhibitors, and hypoxia-activated prodrugs will demonstrate efficacy in the treatment of relapsed MM.

Figure 1

Assessment of drug response in Vk*MYC MM. (A) IC₅₀ (nM) of the indicated drugs obtained by MTT assay in various HMCLs. Each dot represents a cell line, and bars show median values plotted on a log10 scale. In green or red are labeled drugs with single-agent response rates in MM clinical trials above or below 20%, respectively. (B) SPEP and densitometric profile of Vk*MYC mice treated with known active antimyeloma drugs. The day 0 (filled) and day 14 (pink line) traces have been overlaid. The various serum fractions on the densitometry plot are labeled. Arrows point to M-spikes before (day 0) and after (day 14) treatment. (C) The response of Vk*MYC mice to drugs with known (green, red) or (D) unknown (black) activity. Drug activity is quantified by measuring M-spike levels at day 14 after treatment normalized to day 0. Each dot represents an individual M-spike. Bars show median M-spike levels at day 14 normalized to day 0. A blue line in each dot plot arbitrarily separates active drugs (> 50% M-spike reduction, on the left) from inactive ones (< 50% M-spike reduction, on the right). The P value of active drugs compared with vehicle is shown. PLD indicates pegylated liposomal doxorubicin.

Figure 2

Comparison of de novo with transplanted Vk*MYC mice. (Left panel) A flow cytometric analysis of SPL (A) and BM (B) from a matching de novo Vk*MYC and transplant recipient mouse highlighting higher tumor burden and extramedullary MM in tVk*MYC mice. (Middle panel) Immunostaining of (A) SPL and (B) BM sections from de novo and tVk*MYC mice identify a high fraction of proliferating (Ki67⁺, blue) MM (CD138⁺, brown) cells in tVk*MYC mice only. Arrows highlight the rare double-positive cells in the de novo Vk*MYC mice. (B left panel) Cell-cycle analysis on CD138⁺ cells from BM of de novo and tVk*MYC mice. Values (%) ± SD from 3 independent experiments are given.

Figure 3

Characterization of transplanted Vk*MYC mice. (A) Peripheral blood hemoglobin and (B) plasma creatinine levels from tVk*MYC mice compared with nontumor-bearing controls. (C) Renal cast nephropathy, typical of autopsy findings from terminal mice bearing aggressive tVk*MYC MM. (D) Transmission electron microscopy of the glomerular basement membrane (GBM) from a normal C57BL/6 mouse compared at the same magnification to GBM from a tumor-bearing mouse. Note diffuse GBM thickening indicating glomerulopathy. (E) Tibial trabecular bone volume of tVk*MYC myeloma-bearing mice (n = 3) compared with irradiated, age-matched wt controls (n = 3). (F) Histologic evidence of bone lysis in the femoral shaft of a tVk*MYC mouse. (G) Representative micro-CT scans demonstrating cortical lytic lesions in the tibial plateau and fibula from mice bearing the same clone.

Figure 4

Drug activity in tVk*MYC mice. (A) C57BL/6 mice bearing transplanted Vk*MYC tumor (Vk4929 or -3478) were treated with panobinostat alone (4929, n = 16; 3478, n = 6; 25 mg/kg for 4 days followed by 15 mg/kg for 3 weeks) or vehicle control (4929, n = 15; 3478, n = 6; D5W, dextrose water) for a total period of 4 weeks. Serum paraprotein was assessed on day 1 and then weekly for 6 weeks and presented as mean change from levels on day 0 (mean ± SEM). *P < .05 vs Vk4929 vehicle treated. #P < .05 vs Vk3478 vehicle treated. (B) Survival of Vk4929 or -3478 mice treated with panobinostat alone or vehicle (D5W). Median survival of mice treated with panobinostat alone was 36.5 days and 64 days compared with vehicle control-treated mice of 16 days and 32.5 days in Vk4929 or -3478 mice, respectively (P < .05). (C) H&E, acetylated histone H3 and TUNEL-stained BM sections from tVk*MYC 4929 tumor after 8 hours and 12 hours treatment with panobinostat (25 mg/kg) compared with vehicle control (D5W). (D) FACS analysis of viable CD138⁺/B220⁻ PCs in the BM of Vk4929 mice after 8 hours and 12 hours treatment with panobinostat (25 mg/kg) compared with vehicle control (D5W). Values are normalized to the percentage of PCs in vehicle control treated BM (100%). *P < .05. (E) Vk4929 mice were treated with panobinostat (20 mg/kg, days 1-2, reduced to 10 mg/kg onwards to reduce toxicity, n = 7), bortezomib (0.5 mg/kg, twice weekly, n = 7), the combination of both agents (n = 7) or vehicle control (D5W, n = 7) for 4 weeks. Serum paraprotein was assessed on day 1 and then weekly for 6 weeks and presented as mean change from levels on day 0 (mean ± SEM). *P < .05 vs vehicle control. (F) Survival of Vk4929 mice treated with panobinostat, bortezomib, the combination of both agents and vehicle control (D5W). Median survival of mice were as follows: panobinostat alone 32 days; bortezomib alone 24 days; the combination of both agents 36 days; vehicle control-treated mice 18 days. (G) FACS analysis of viable CD138⁺/B220⁻ PCs in the BM of Vk4929 mice after 5 days of treatment with panobinostat (10 mg/kg daily), bortezomib (0.5 mg/kg, days 1 and 4), the combination of both agents or vehicle control (D5W). Values are normalized to the percentage of PCs in vehicle control treated BM (100%). *P < .05 vs vehicle control.

Figure 5

Establishment and characterization of bortezomib-resistant Vk*MYC tumors. (A) M-spike levels (normalized to day 0) of 2 mice (2917 and 2777) treated with bortezomib at 0.5 mg/kg (gray and black lines) or 1 mouse (B763) at 1 mg/kg (dotted line) for 2 weeks (1 cycle). Arrowheads indicate the time of administration of 1 cycle of therapy, their color refers to the treated mouse and their size is proportional to the drug dosage (medium = 0.5 mg/kg, large = 1 mg/kg). (B) M-spike levels (normalized to day 0) of 3 mice (2825, B1165, and 3569) treated with continuous escalating doses of bortezomib. Arrowheads and their size indicate treatment: small = 0.16 mg/kg, medium = 0.25 mg/kg, large = 0.5 mg/kg, and their color refers to the treated mouse. (C-D) Response to single or combination treatment of 2 independent bortezomib-resistant Vk*MYC tumors (Vk12598 and Vk12653) propagated by serial transplantation into congenic wild-type (wt) mice. M-spike levels at day 14 normalized to day 0 are shown in log2 scale. Each dot represents a mouse. Bars show median day 14 M-spike levels normalized to day 0. The P value of active drugs compared with vehicle is shown. (E) SPEP performed on the Vk12598 line of bortezomib-resistant mice shows effective bortezomib + HDACi combination treatment but ineffective single-agent activity of these 2 drug. Arrows point to M-spikes.