SUPPLEMENTARY APPENDIX

Phase I Trial of the Combination of Flavopiridol and Imatinib mesylate in Patients with Bcr-Abl+ Hematological Malignancies

PHARMACODYNAMIC STUDIES: STAT5 ANALYSES BY FLOW CYTOMETRY

Methods

Blood samples were received from University Hospitals of Cleveland (CWRU) immediately after withdrawal. Samples from Johns Hopkins University (JHU), Virginia Commonwealth University (VCU), and University of Pennsylvania (UPN) were received 24 hours after withdrawal by commercial overnight carrier at room temperature. Upon receipt, nucleated cells were counted with a Coulter Counter (Beckman Coulter, Miami, FL) or Guava cytometer (Millipore, Billerica, MA). Aliquots of two million leukocytes were delivered to 3 tubes, which were incubated for 30 min at room temperature in the dark (RTD) with a cocktail containing CD34-PC7 and CD45-PerCp (BD Biosciences, San Jose, CA); CD11b-PE and CD15-FITC (Immunotech, Vaudreuil-Dorion, Quebec), and 25ug mouse IgG (Invitrogen, Carlsbad, CA). Antibodies were used at manufacturer's recommended quantity. After incubation, 1 ml phosphate buffered saline (PBS) with 2% fetal bovine serum (PBS/FBS) was added to each tube and centrifuged at low speed; supernatants removed and 1 ml of PermiFlow (Invirion Diagnostics) was added. The tubes were incubated for 40 min RTD then centrifuged and resuspended in 1 ml PBS, incubated for 10 min, then washed twice with 1 ml FBS/PBS. In the washes post PermiFlow, the red blood cells lysed. Two tubes were then incubated with 20ul antibody reactive with phospho-Y694-Stat5, conjugated with AlexaFluor 647 (pStat5, BD Biosciences) and 30 ul FBS/PBS for 30 min RTD, then washed twice in 1 ml FBS/PBS. After a final centrifugation, cells were resuspended in 650 ul PBS with 1 ug DAPI. Measurements were made on an LSR II equipped with violet, blue, and red lasers (BD Biosciences). The tube without pStat5 was used to check for pStat5 staining. During the assay workup, staining of CML samples were compared to isotype controls for pStat5, samples treated with Imatinib, and blood from healthy donors. Analysis was performed offline with WinList 7.0 (Verity Software House, Topsham, ME) as described in Figure S1.

Figure S1a. Analytical Scheme. The example is a patient with an abnormally high pStat5+ blast cell percentage at baseline. Peripheral blood shows normal features (lymphocytes, monocytes, and granulocytes) and in addition, abnormal features of CML (high levels of CD34+ blasts and immature granulocytes, basophils, nucleated erythrocytes. A: pretreatment pattern for this patient. B: gating of the CD34+ blast cells (pretreatment sample). C: pretreatment dot plot with color coding for blasts (red), blast pStat5-positive cells (purple), and blast S+G2+M cells (green). Note that pStat5-positive and S+G2+M cells are predominantly in the more granulated CD34+ cells. D, F, H cell cycle pattern with G1 cells identified by red or purple colors and S+G2+M identified by green color for pretreatment (D), post-Imatinib (F), and post-flavopiridol (H) samples. The analogous pStat5 profiles are displayed in E, G, and I with the pStat5-positive region defined by the color purple. The fraction of pStat5-positive cells in G is less than in E. This behavior was taken to be evidence of Imatinib activity detection. The fraction of S+G2+M cells in H is greater than the fraction of S+G2+M cells in F. This behavior is interpreted as evidence for failure to detect flavopiridol activity.

Figure S1b. Comparison of normal and CML. Panels A-E are equivalent to A-E in Figure S1a except that the sample comes from a healthy person. Panels d and e are replicated from Figure S1a for comparison of a CML patient to a healthy individual. Note the low numbers of circulating CD34+ cells from the blast region (red dots in panel C); apparent absence of S phase or appreciable G1 pStat5+ cells. False positive pStat5 frequency is 0.3%.

Figure S2. Expected and observed values for pStat5+ (%) and S+G2+M (%). Shown are the patients that displayed pStat5+ positive cell percentages 2 standard deviations above healthy volunteers. We expected that percentages of pStat5+ and S+G2+M blasts would decrease after Imatinib treatment. We expected flavopiridol to either maintain a G1 arrest or increase the fraction of G1 cells (decrease S+G2+M). Patients 9 and 15 performed to expectations. All patients displayed a decrease in pStat5 positivity, although the shift for Patient 9 was not significant.

PHARMACOKINETIC ANALYSES

Blood samples for imatinib and flavopiridol pharmacokinetics were collected over 49 hours during days 1 and 2 of the first cycle. Sample times were: before treatment, and at 1, 2, 4, 8, 12 and 24 hours after the day 1 dose; and at the same times around the day 2 imatinib dose, which was taken 25 hours after the first dose, at the end of the 1 h infusion of flavopiridol. Plasma was prepared by centrifugation and immediately frozen at -20 ºC. Plasma concentrations of imatinib and its active metabolite, CGP74588, were determined with a validated liquid chromatographic-mass spectrometric (LC-MS) assay [1]. Plasma concentrations of flavopiridol were determined using a validated high performance liquid chromatography (HPLC) assay [2] with the lower limit of quantification (LLOQ) being 20 ng/ml. Two individuals with insufficient data points (2-3 samples) were excluded from the flavopiridol pharmacokinetic analysis. Calculated parameters included the area under the plasma concentration-time curve (AUC), AUC extrapolated to infinity (AUCinf), clearance (CL), volume of distribution at steady-state (Vss), and terminal half-life (t1/2). The peak flavopiridol concentration (Cmax) was determined as the concentration achieved at the end of the 1-hr infusion. Effects, if any, of flavopiridol on imatinib pharmacokinetics were assessed. Given the reported half-life of imatinib and its metabolite of 18 and 40 h, respectively [3], imatinib concentrations at 24 and 25 hours (before and after the infusion of flavopiridol) would be expected to be virtually identical. Statistical analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, IL) and Prism 5 (GraphPad Software, La Jolla, CA) for imatinib and flavopiridol pharmacokinetics, respectively. A two-tailed paired t-test was used for all comparisons, and a p value < 0.05 was considered statistically significant.

Figure S3. Individual (A-C) and mean (D) pharmacokinetic profiles of flavopiridol following a 1-hr infusion with either 400 mg/day (gray) or 600 mg/day (black) of imatinib in cycle 1.

Table S1. Pharmacokinetic parameters of flavopiridol at various dose levels.

Dose
Level / Flavopiridol
(mg/m2) / Imatinib
(mg/day) / Cmax
(ng/ml) / AUCinf
(ng/ml·hr) / CL
(L/hr) / Vss
(L) / t1/2
(hr)
1 (n=3) / 30 / 400 / 969±255 / 2314±828 / 29.9±15.1 / 63.2±30.8 / 2.07±0.78
2 (n=2) / 30 / 600 / 1584±380 / 3075±1202 / 24.6±7.4 / 60.0±37.1 / 2.60±2.30
3 (n=3) / 45 / 600 / 2165±726 / 4246±549 / 19.8±4.1 / 66.5±58.9 / 2.28±0.80
4 (n=4) / 60 / 600 / 1790±684 / 4003±955 / 27.3±8.2 / 73.4±26.3 / 2.44±0.86

Data are presented as mean ± SD. Abbreviations: Cmax, peak concentration; AUCinf, area under the plasma concentration-time curve extrapolated to infinity; CL, clearance; Vss, volume of distribution at steady-state; t1/2, terminal half-life.

REFERENCES

1.  Parise RA, Ramanathan RK, Hayes MJ, Egorin MJ. Liquid chromatographic- mass spectrometric assay for quantitation of imatinib and its main metabolite (CGP 74588) in plasma. J Chromatogr B Analyt Technol Biomed Life Sci 2003; 791: 39-44.

2.  Zhai S,Sausville E, Figg WD. A high-performance liquid chromatography method using ultraviolet detection for the quantitation of flavopiridol from human plasma. Biomed Chromatogr, 2002; 16(6): 379-82.

3.  Peng B, Lloyd P, Schran H. Clinical pharmacokinetics of imatinib. Clin Pharmacokinet 2005; 44(9): 879-94.

7