Shrey Ramnath

The patient-derived cells have numerous genomic amplifications and deletions, consistent with derivation of the cells from the osteosarcoma itself, rather than from bystander cells

Sarcospheres from a subset of patients are sensitive to romidepsin. Metabolic activity of sarcospheres from 18 human (A) and 10 canine (B) patients was measured after exposure for 48 hours to the indicated concentrations of romidepsin. Black curves represent sensitive subset of patients with ED50s 1,000 nmol/L. Each curve represents the mean of at least n = 3 independent experiments with sarcospheres from an individual patient sample. Each experiment included six sarcospheres at each of the indicated doses. Asterisks (*) indicate dose groups (100 and 1,000 nmol/L) that are significantly different (P < 0.05) from the vehicle-treated group for all sensitive samples based on one-way ANOVA with Dunnett multiple comparisons tests. C, Diameters of human patient–derived sarcospheres were measured every 2 hours during exposure to the indicated concentrations of romidepsin or vehicle by the Incucyte S3 Live Cell Analysis System (Sartorius). Each symbol represents mean ± SEM of N = 6 patients (two romidepsin sensitive and four romidepsin resistant). Sarcospheres from each patient were analyzed in 1–3 independent experiments. Each experiment included 3–6 sarcospheres per dose group. D, Representative images of romidepsin-sensitive sarcospheres from experiments depicted in C. Yellow size bars in lower left of images show 600 mm. Time stamp is shown in lower right of each image. E, Representative images from romidepsin-resistant patients from experiments depicted in C. Yellow size bars in lower left of images show 800 mm. Time stamp is shown in lower right of each image.

Clinical information for patients and romidepsin ED50s.

Dose responses and ED50s from secondary screen

Toxicity and therapeutic indices on Normal Human Osteoblasts

Suppliers and Storage Conditions for 13 Drugs in Secondary Screen

The primary effects of romidepsin are a transient block in proliferation for 143B sarcospheres and cytotoxicity for LM7 sarcospheres. 143B (A–D) and LM7 (E–H) sarcospheres were exposed to the indicated concentrations of romidepsin or vehicle for 24 hours (A, B, E, and F) or the indicated times (C, D, G, and H). A and E, Cell cycles were analyzed by flow cytometry after exposure to 10 nmol/L romidepsin or vehicle for 24 hours. Each symbol represents the percentage of the cell population in each phase from an individual experiment with 30 sarcospheres per group. Colored bars show means ± SD of N = 3–4 independent experiments. A one-way ANOVA with a Dunnett multiple comparisons test was used to compare experimental groups with the vehicle control group *, P < 0.05; ***, P < 0.01; ****, P < 0.001. B and F, Recovery from a 24-hour romidepsin exposure was assessed by repeated measurements of metabolic activity on days 2, 4, 8, and 14 after romidepsin addition. All repeated measurements were on the same set of sarcospheres, except that measurements on day 0 were on a separate plate of sarcospheres. Each symbol represents mean ± SD of N = 3 independent experiments. Each experiment included six sarcospheres per dose group. Friedman one-way repeated measures ANOVA by ranks with a Dunn multiple comparisons test was used to compare each experimental group with the vehicle control group over time. ***, P < 0.001. C and G, Sarcosphere diameters were measured every 2 hours during exposure to the indicated concentrations of romidepsin or vehicle by the Incucyte S3 Live Cell Analysis System (Sartorius). Each symbol represents mean ± SEM of N = 3–4 independent experiments. Each experiment included 3–6 sarcospheres per dose group. A Friedman test with a Dunn multiple comparisons test was used to compare each experimental group with the vehicle control group for 143B sarcosphere size experiments: *, P < 0.05 (1 nmol/L group), ***, P < 0.001 (10–10,000 nmol/L groups). D and H, Representative images from experiments depicted in C and G. Yellow size bars on the lower left of images show 800 mm. Timestamp is indicated in the lower right of each image.

Romidepsin is more promising than all other tested HDIs. Romidepsin was compared with the three other FDA-approved HDIs, as well as eight that are in clinical trials. Five of the HDIs in clinical trials (abexinostat, givinostat, mocetinostat, pracinostat, and quisinostat) weakly inhibited sarcosphere metabolic activity when tested at their Cmax in preliminary studies and were therefore excluded from further study (Supplementary Fig. S7). All heatmaps in this figure are sorted by the overall row means shown in the first column of each heat map. Dark blue indicates highest values (most favorable), dark red the lowest values (least favorable), and white is set at the median of the row means. A and B, Ratios of Cmax/sarcosphere (Sarco) ED50 with and without MAP reflect potency at clinically achievable concentrations. Cmax and dose responses are shown in Supplementary Table S3 and Supplementary Fig. S8. ED50s were determined for sarcospheres derived from LM7, 143B, and MG63.3-GFP cell lines in N = 3 independent dose–response experiments with each drug. Each experiment included three sarcospheres per group. C, Ratios of TD50/sarcosphere (Sarco) ED50 reflect therapeutic indices with NHOst. TD50s were determined in N = 3 independent dose–response experiments with each drug. Each experiment included at least three culture wells per group.

Dose responses and Bliss Prediction of Additivity for 143B sarcospheres in secondary screen

Romidepsin has the largest overall therapeutic index of all 13 hits from the initial screen. Data represent dose–response experiments for top 13 hits from initial screen. Ratios of TD50/sarcosphere (Sarco) ED50 reflect therapeutic indices with NHOst and non-transformed human SAEC. White on the color scale is set at the median of the overall row means, shown in the first column. TD50s, doses that reduce metabolic activity of NHOst and SAEC by 50%, are shown in Supplementary Figs. S5A and S6A. The TD50/ED50 ratios are shown in Supplementary Figs. S5B–S5D and S6B–S6D. TD50s were determined in N = 3 independent dose–response experiments with each drug. Each experiment included at least three culture wells per group. ED50s, doses that reduce metabolic activity of 143B, LM7, or MG63.3-GFP sarcospheres by 50%, are shown in Supplementary Fig. S1B–S1D and Supplementary Table S3.