Skip to main content
NCBI home page
ACS Medicinal Chemistry Letters logoLink to ACS Medicinal Chemistry Letters
. 2024 Apr 4;15(5):739–745. doi: 10.1021/acsmedchemlett.4c00071

Intracellular Pharmacokinetics of Activated Drugs in a Prodrug–Enzyme–Ultrasound System: Evaluations on ZD2767P+CPG2+US

Tinghe Yu †,*, Xinya Li †, Tianran Yu ‡, Mengjie Chen †, Yong Sun §, Rui Ran †
PMCID: PMC11089658  PMID: 38746880

Abstract

graphic file with name ml4c00071_0002.jpg

Intracellular pharmacokinetics (PK) of activated drugs is a window to understanding the pharmacodynamics of prodrug–enzyme–ultrasound therapy. Herein PK of ZD2767D (i.e., activated drug) in the ZD2767P+CPG2+US system on A549, A549/DDP, SKOV3, and SKOV3/DDP cells were evaluated (A549/DDP and SKOV3/DDP were cisplatin-resistant sublines). The noncompartment model under extravascular input mode was deemed appropriate for evaluating drug level vs time curves; Cmax, AUClast, MRTlast, Vz, and Cl can reflect the PK feature, but t1/2, AUCinf, and MRTinf were irrational; higher accumulation and slower elimination characterized the PK mechanism of ZD2767P+CPG2+US; enhanced permeability and retention effect can be assessed with Cmax, AUClast, MRTlast, and tlast; ultrasound equivalently modulated Cmax and AUClast in sensitive and resistant cells. The experimental design and dose proportionality were discussed.

Keywords: Intracellular pharmacokinetics, Prodrug−enzyme−ultrasound therapy, ZD2767D, Experimental design, Pharmacokinetics model, Therapeutic implications

Carboxypeptidase G2 (CPG2) hydrolyzes ZD2767P (4-[N,N-bis(2-iodoethyl)amino]phenoxycarbonyl-l-glutamic acid) to release the cytotoxic moiety ZD2767D (4-[N,N-bis(2-iodoethyl)amino]phenol), i.e., prodrug–enzyme therapy. The ZD2767P+CPG2 system realizes a targeted cancer treatment but is clinically limited by low efficacy and toxicity primarily due to off-target activation of ZD2767P.1

Ultrasound (US) can be used to permeabilize vessels and cell membranes to enhance drug uptake (i.e., sonochemotherapy).2 Ultrasound can improve the CPG2 activity to increase the early yield of ZD2767D, with the same final output.3 We therefore design the ZD2767P+CPG2+US system, which can be considered the combination of prodrug–enzyme therapy and sonochemotherapy.4 This modality has been tested on cisplatin-sensitive and -resistant lung and ovarian cancers, given that cisplatin is the first-line agent and resistance has been a therapeutic challenge.5,6 Preclinical trials demonstrated that the ZD2767P+CPG2+US system improved efficacy and reduced toxicity in comparison with the ZD2767P+CPG2 system: a smaller tumor volume and a longer survival time were noted.3,7

Intracellular pharmacokinetics (PK) of the activated drug determines pharmacodynamics of a prodrug–enzyme system but is limitedly understood.4 Cisplatin resistance associated with intracellular PK is the pretarget mechanism: reduced influx and/or increased efflux decrease the amount of intracellular platinum; the PK mechanism varies between cancer and cell types and cannot be overcome by increasing the cisplatin dose.5,6 A drug with a different PK profile may bypass the intracellular PK mechanism of cisplatin resistance. Enhanced permeability and retention effect (EPR) can be an intracellular PK mechanism of sonochemotherapy, which increases the amount of drugs entering the cell and has been detected in certain resistant cancer cells.2,8−12

ZD2767D has a short half-life (t1/2; ≈2 min in plasma) thus being an ideal model drug for exploring intracellular PK: a prolonged retention time is easily observable.13 ZD2767D is generated only in the extracellular medium since CPG2 cannot cross the cell membrane, and enters the cell via passive diffusion.3 Thus, the ZD2767P+CPG2+US system can be used to explore the intracellular PK of activated drug. Here we investigated the intracellular PK of ZD2767D in cisplatin-sensitive and -resistant cancer cell lines.

Two pairs of sensitive-resistant cell lines (A549 and A549/DDP, and SKOV3 and SKOV3/DDP) were used to improve clinical relevance. The ZD2767D level vs time (Ct vs t) curves were illustrated in Figures 1A–D and S1–S3. The drug levels in group ZD2767P+CPG2+US were higher than those in group ZD2767P+CPG2. Noticeably, the level plummeted to 0 (i.e., below the detection limit) after washing out drugs. Posthoc power was ≥0.80 in 81/82 items with statistical significance, confirming high statistical power (Tables 1, 2).14

Figure 1.

Figure 1

Open in a new tab

Typical Ct vs t curves of ZD2767D (A–D): extracellular drugs were washed out after 60 min. Dose proportionality was assessed with Cmax and AUClast using the linear regression (E–H) or power (I–L) model: the fitted line in the linear regression model indicated proportionality; in the power model, the bar was the 90% CI of β, and the gray area illustrated the theoretical limit of β value. Columns from left to right: A549, A549/DDP, SKOV3, and SKOV3/DDP cells.

Table 1. PK Parameters of ZD2767D in A549 and A549/DDP Cells (n = 6–9)a,b.

  100 μM
 200 μM
 400 μM
  ZD2767P+CPG2 ZD2767P+CPG2+US ZD2767P+CPG2 ZD2767P+CPG2+US ZD2767P+CPG2 ZD2767P+CPG2+US
A549            
Cmax (nmol/mg) 3.89 ± 0.12 7.42 ± 0.81c 6.16 ± 0.77 10.82 ± 0.57c 13.34 ± 0.25 20.88 ± 0.63c
  power = 1.0000   power = 1.0000   power = 1.0000
Clast/Cmax 0.85 ± 0.06 0.71 ± 0.06 0.39 ± 0.01 0.67 ± 0.04 0.33 ± 0.03 0.92 ± 0.03
tmax (min) 5 5 5 5 15 15
t1/2 (min) 132.31 ± 60.19 563.78 ± 537.28 36.45 ± 3.73 210.22 ± 36.34c 26.81 ± 2.32 429.73 ± 163.80c
  power = 0.6680   power = 1.0000   power = 1.0000
AUClast ((nmol/mg) min) 101.93 ± 4.33 343.62 ± 25.78c 268.16 ± 29.27 484.93 ± 17.85c 521.45 ± 19.99 1084.85 ± 12.22c
  power = 1.0000   power = 1.0000   power = 1.0000
AUCinf ((nmol/mg) min) 755.59 ± 354.66 4545.84 ± 3945.71 392.29 ± 33.16 2707.14 ± 485.51c 691.23 ± 54.37 13029.17 ± 4613.54c
  power = 0.8187   power = 1.0000   power = 1.0000
AUClast/AUCinf 0.16 ± 0.07 0.12 ± 0.07 0.68 ± 0.02 0.18 ± 0.03 0.76 ± 0.03 0.09 ± 0.04
MRTlast (min) 15.01 ± 0.24 28.66 ± 0.28c 25.39 ± 0.29 29.04 ± 0.31c 27.24 ± 0.56 32.50 ± 0.15c
  power = 1.0000   power = 1.0000   power = 1.0000
MRTinf (min) 192.59 ± 87.15 812.15 ± 774.70 53.18 ± 4.04 303.27 ± 52.68c 44.72 ± 3.43 626.22 ± 235.95c
  power = 0.6643   power = 1.0000   power = 1.0000
Vz (μM/(nmol/mg)) 25.38 ± 0.55 16.80 ± 1.89c 27.16 ± 4.98 22.43 ± 0.88 22.37 ± 0.19 18.86 ± 0.62c
  power = 1.0000   power = 0.8012   power = 1.0000
Cl (μM/(nmol/mg)/min) 0.16 ± 0.07 0.04 ± 0.02c 0.51 ± 0.05 0.08 ± 0.01c 0.58 ± 0.05 0.03 ± 0.01c
  power = 0.9986   power = 1.0000   power = 1.0000
A549/DDP            
Cmax (nmol/mg) 3.96 ± 0.38 7.82 ± 0.97c 6.29 ± 0.49 11.23 ± 0.76c 12.47 ± 0.76 20.72 ± 0.40c
  power = 1.0000   power = 1.0000   power = 1.0000
Clast/Cmax 0.82 ± 0.08 0.52 ± 0.37 0.38 ± 0.09 0.66 ± 0.01 0.33 ± 0.01 0.92 ± 0.00
tmax (min) 5 5 5 5 15 15
t1/2 (min) 129.12 ± 107.55 372.63 ± 271.34 34.26 ± 8.56 238.94 ± 9.62c 27.18 ± 0.48 377.68 ± 21.80c
  power = 0.7064   power = 1.0000   power = 1.0000
AUClast ((nmol/mg) min) 104.14 ± 11.16 400.04 ± 71.76c 277.39 ± 34.30 489.73 ± 36.36c 487.74 ± 27.80 1067.84 ± 19.49c
  power = 1.0000   power = 1.0000   power = 1.0000
AUCinf ((nmol/mg) min) 729.14 ± 540.00 3341.91 ± 2199.63 403.02 ± 95.18 3058.42 ± 210.05c 649.33 ± 33.71 11408.97 ± 563.89c
  power = 0.9333   power = 1.0000   power = 1.0000
AUClast/AUCinf 0.20 ± 0.09 0.38 ± 0.45 0.70 ± 0.09 0.16 ± 0.01 0.75 ± 0.01 0.09 ± 0.01
MRTlast (min) 14.91 ± 0.30 30.95 ± 2.77c 25.25 ± 1.02 28.97 ± 0.10c 27.27 ± 0.17 32.43 ± 0.12c
  power = 1.0000   power = 1.0000   power = 1.0000
MRTinf (min) 187.63 ± 155.07 539.22 ± 387.35 51.03 ± 12.07 344.61 ± 13.90c 45.19 ± 0.49 551.32 ± 31.40c
  power = 0.7150   power = 1.0000   power = 1.0000
Vz (μM/(nmol/mg)) 24.87 ± 2.54 13.04 ± 5.53 24.60 ± 3.17 22.64 ± 1.78c 24.22 ± 1.59 19.10 ± 0.41c
  power = 0.9999   power = 0.3661   power = 1.0000
Cl (μM/(nmol/mg)/min) 0.19 ± 0.11 0.08 ± 0.09 0.52 ± 0.10 0.07 ± 0.00c 0.62 ± 0.03 0.04 ± 0.00c
  power = 0.6413   power = 1.0000   power = 1.0000
Open in a new tab
a

Cmax: peak level; Clast: last measurable level; tmax: time of Cmax; t1/2: half-life; AUClast: area under Ct vs t curve from zero to last measurable level; AUCinf: AUC from zero to infinity; MRTlast: mean residence time from zero to last measurable level; MRTinf: MRT from zero to infinity; Vz: volume of distribution; Cl: clearance.

b

Clast/Cmax, tmax, and AUClast/AUCinf were not statistically compared.

c

vs ZD2767P+CPG2, p < 0.05.

Table 2. PK Parameters of ZD2767D in SKOV3 and SKOV3/DDP Cells (n = 6–9)a.

  100 μM
 200 μM
 400 μM
  ZD2767P+CPG2 ZD2767P+CPG2+US ZD2767P+CPG2 ZD2767P+CPG2+US ZD2767P+CPG2 ZD2767P+CPG2+US
SKOV3            
Cmax (nmol/mg) 3.63 ± 0.22 6.87 ± 0.44b 6.41 ± 0.68 10.76 ± 0.39b 11.99 ± 0.41 19.75 ± 1.25b
  power = 1.0000   power = 1.0000   power = 1.0000
Clast/Cmax 0.69 ± 0.17 0.57 ± 0.09 0.33 ± 0.02 0.69 ± 0.01 0.37 ± 0.02 0.91 ± 0.01
tmax (min) 5 5 5 5 15 15
t1/2 (min) 58.34 ± 28.08 108.43 ± 52.84 29.26 ± 1.83 231.40 ± 41.63b 30.23 ± 1.39 339.91 ± 39.39b
  power = 0.7093   power = 1.0000   power = 1.0000
AUClast ((nmol/mg) min) 91.06 ± 10.40 298.38 ± 12.59b 272.20 ± 23.86 488.14 ± 15.18b 489.56 ± 27.48 1016.84 ± 56.12b
  power = 1.0000   power = 1.0000   power = 1.0000
AUCinf ((nmol/mg) min) 326.22 ± 150.48 934.35 ± 387.64b 362.98 ± 37.43 2978.78 ± 516.61b 684.64 ± 52.57 9872.20 ± 1645.82b
  power = 0.9924   power = 1.0000   power = 1.0000
AUClast/AUCinf 0.36 ± 0.19 0.35 ± 0.10 0.75 ± 0.02 0.17 ± 0.03 0.72 ± 0.02 0.10 ± 0.01
MRTlast (min) 14.31 ± 0.86 27.18 ± 1.01b 24.65 ± 0.25 29.18 ± 0.12b 27.62 ± 0.41 32.36 ± 0.31b
  power = 1.0000   power = 1.0000   power = 1.0000
MRTinf (min) 85.18 ± 41.00 155.72 ± 76.22 44.00 ± 2.50 334.02 ± 59.75b 49.22 ± 2.05 496.73 ± 57.27b
  power = 0.6862   power = 1.0000   power = 1.0000
Vz (μM/(nmol/mg)) 25.52 ± 1.26 16.46 ± 1.25b 23.39 ± 1.75 22.39 ± 0.61 25.53 ± 0.87 20.02 ± 1.18b
  power = 1.0000   power = 0.3666   power = 1.0000
Cl (μM/(nmol/mg)/min) 0.42 ± 0.27 0.12 ± 0.03 0.56 ± 0.06 0.07 ± 0.01b 0.59 ± 0.05 0.04 ± 0.01b
  power = 0.9120   power = 1.0000   power = 1.0000
SKOV3/DDP            
Cmax (nmol/mg) 4.12 ± 0.38 8.07 ± 0.35b 7.59 ± 0.89 12.32 ± 0.14b 12.80 ± 0.44 22.43 ± 0.58b
  power = 1.0000   power = 1.0000   power = 1.0000
Clast/Cmax 0.80 ± 0.03 0.71 ± 0.03 0.27 ± 0.06 0.68 ± 0.04 0.29 ± 0.00 0.87 ± 0.04
tmax (min) 5 5 5 5 15 15
t1/2 (min) 82.45 ± 15.18 286.09 ± 132.39b 28.32 ± 6.77 241.39 ± 130.77b 24.14 ± 0.31 247.58 ± 74.13b
  power = 0.9957   power = 0.9983   power = 1.0000
AUClast ((nmol/mg) min) 103.13 ± 7.93 366.83 ± 9.22b 293.33 ± 30.41 548.82 ± 21.65b 495.57 ± 14.18 1127.57 ± 46.01b
  power = 1.0000   power = 1.0000   power = 1.0000
AUCinf ((nmol/mg) min) 489.55 ± 49.15 2716.16 ± 1058.50b 381.23 ± 51.08 3523.72 ± 1805.03b 624.62 ± 19.27 8166.25 ± 2476.16b
  power = 1.0000   power = 0.9994   power = 1.0000
AUClast/AUCinf 0.21 ± 0.03 0.16 ± 0.06 0.78 ± 0.08 0.18 ± 0.05 0.79 ± 0.00 0.15 ± 0.04
MRTlast (min) 14.86 ± 0.13 28.76 ± 0.18b 23.90 ± 0.82 29.15 ± 0.43b 26.68 ± 0.08 32.06 ± 0.11b
power = 1.0000   power = 1.0000   power = 1.0000
MRTinf (min) 120.96 ± 21.87 413.70 ± 188.72b 41.80 ± 9.19 349.15 ± 187.73b 40.75 ± 0.43 363.70 ± 106.40b
  power = 0.9961   power = 0.9984   power = 1.0000
Vz (μM/(nmol/mg)) 24.17 ± 2.41 14.77 ± 1.27b 21.48 ± 4.29 19.59 ± 1.39 22.32 ± 0.66 17.51 ± 0.43b
  power = 1.0000   power = 0.2418   power = 1.0000
Cl (μM/(nmol/mg)/min) 0.21 ± 0.02 0.04 ± 0.02b 0.53 ± 0.07 0.06 ± 0.02b 0.64 ± 0.02 0.05 ± 0.02b
  power = 1.0000   power = 1.0000   power = 1.0000
Open in a new tab
a

Clast/Cmax, tmax, and AUClast/AUCinf were not statistically compared.

b

vs ZD2767P+CPG2, p < 0.05.

Ct vs t curves at 100 μM ZD2767P were fitted with multiple PK models to determine appropriate models. Non- (extravascular input, intravenous bolus injection, and intravenous infusion modes) and one-compartment (extravascular input mode) models could evaluate curves in groups ZD2767P+CPG2 and ZD2767P+CPG2+US (Table S1). Then, t1/2 and MRTlast (mean residence time from zero to last measurable level) were compared with tlast (time of last measurable level) to determine the rationality of candidate models. In the noncompartment model, analyses under extravascular input and intravenous injection modes were equivalent, and MRTlast was <tlast (i.e., logical). In other modes, t1/2 and MRT were much longer than tlast or <0 (i.e., illogical) (Table S2). Therefore, the noncompartment model under extravascular input mode was the appropriate PK model.

A compartment model requires specific assumptions (i.e., a mathematic function) and a minimum number of data points.15 However, not all of these criteria can be met in a cell PK trial. Each data point was obtained from a distinct cell population, increasing the heterogeneity. The noncompartment model was advantageous because it required few premises, given the nonlinear behavior of therapeutic ultrasound. The noncompartment model utilized the trapezoid rule, thereby mirroring the experimental settings.16

Cellular uptake of fluorescent dye SYTOX green under insonation was described using the two- or three-compartment model in previous investigations.17−19 The present data on ZD2767D indicated that compartment models cannot outline the intracellular PK of a cytotoxic drug. This discrepancy prompted concerns, since fluorescent molecules were widely used as mimetics to trace the kinetics of drug uptake in sonochemotherapy.20

Ct vs t curves were evaluated using the noncompartment model under extravascular input mode. In 4 cell lines, higher Cmax (peak level), AUClast (area under Ct vs t curve from zero to last measurable level) and MRTlast, and lower Vz (volume of distribution) and Cl (clearance) were detected in group ZD2767P+CPG2+US in comparison with group ZD2767P+CPG2 (Tables 1, 2). These data demonstrated that higher accumulation and slower elimination characterized the PK mechanism of ZD2767P+CPG2+US. These PK parameters were deduced from only measured values from zero to tlast, thereby generalizing the PK features.

Cmax and AUClast determine the efficacy of a drug; the ratio of the geometric mean is a means to balance the considerable background differences in comparisons of PK.21 Ratios (ZD2767P+CPG2+US / ZD2767P+CPG2) in A549 and A549/DDP cells, and in SKOV3 and SKOV3/DDP cells were nearly equal, indicating that ultrasound equivalently modulated PK in sensitive and resistant cells (Table 3). That the PK property of ZD2767D differed from that of cisplatin was a mechanism of defeating cisplatin resistance by ZD2767P+CPG2+US.3,5,7

Table 3. Ratios of Geometric Means of Cmax and AUClasta.

  100 μM
 200 μM
 400 μM
  Cmax AUClast Cmax AUClast Cmax AUClast
A549 1.47 (1.41, 1.54) 1.26 (1.25, 1.28) 1.32 (1.27, 1.37) 1.11 (1.10, 1.12) 1.17 (1.17, 1.18) 1.12 (1.11, 1.12)
A549/DDP 1.50 (1.40, 1.61) 1.29 (1.26, 1.32) 1.32 (1.30, 1.33) 1.10 (1.09, 1.11) 1.20 (1.18, 1.22) 1.13 (1.12, 1.13)
SKOV3 1.50 (1.44, 1.56) 1.26 (1.25, 1.28) 1.28 (1.25, 1.32) 1.10 (1.10, 1.11) 1.20 (1.19, 1.21) 1.12 (1.12, 1.12)
SKOV3/DDP 1.48 (1.43, 1.54) 1.27 (1.26, 1.28) 1.25 (1.20, 1.30) 1.11 (1.10, 1.13) 1.22 (1.21, 1.23) 1.13 (1.13, 1.13)
Open in a new tab
a

ZD2767P+CPG2+US / ZD2767P+CPG2; 90% CI; n = 6–9

Higher standard deviations of t1/2, AUCinf (AUC from zero to infinity), and MRTinf (MRT from zero to infinity) demonstrated greater heterogeneity. Clast (last measurable level) decided λz (slope of the terminal phase) and then determined t1/2 (t1/2 = ln[2]/λz). AUCinf (eq 1) and MRTinf (eq 2)

graphic file with name ml4c00071_m001.jpg 1
graphic file with name ml4c00071_m002.jpg 2

(Ci and ti formed one base of a trapezoid, Ci+1 and ti+1 formed the other base, and Δt = ti+1 – ti) included extrapolated values from tlast to infinity, thereby increasing variances.16 These data were consistent with the drastic variances observed in Clast/Cmax and AUClast/AUCinf. Further, these PK parameters lacked logical consistencies with an increasing dose (Tables 1, 2). Thus, t1/2, AUCinf, and MRTinf were inappropriate parameters to describe the PK feature.

EPR can be understood from the perspective of intracellular PK. Higher Cmax and AUClast values indicated enhanced permeability, and a higher AUClast and a longer MRTlast represented an enhanced retention effect. At 100 μM ZD2767P, tlast was 30 and ≥60 min in groups ZD2767P+CPG2 and ZD2767P+CPG2+US, respectively (Figure S1). Thus, a longer tlast indicated an enhanced retention effect. Extra- and intra-cellular drug levels rapidly reached equilibrium at a low dose, i.e., a short tlast in group ZD2767P+CPG2. Insonation prolonged the duration of drug influx, leading to a longer tlast in group ZD2767P+CPG2+US. Equilibrium required a longer time at a higher dose, masking the effect of insonation; therefore, tlast cannot reflect the enhanced retention effect at higher doses.

The dose proportionality indicates that Cmax and AUClast at any dose can be predicted using available PK parameters, i.e., adjusting the drug dose.22 Here 3 doses were set, and thus proportionality was assessed. The linear regression model utilizing AUClast demonstrated proportionality in A549/DDP, SKOV3, and SKOV3/DDP cells in group ZD2767P+CPG2, and in A549 cells in group ZD2767P+CPG2+US (Figure 1E–H, Table S3). These data indicated that proportionality was cell-type dependent.

The verdict using AUClast has better therapeutic relevancy in comparison with Cmax; the rigorness is power model > quadratic linear regression model > linear regression model.22 The power model had better clinical relevancy but required more data points. Thus, the power model should be tried when having ≥20 data points. Here no solid proportionality was confirmed with the power model; PK in SKOV3 cells in group ZD2767P+CPG2+US can approximate proportionality since the confidence interval (CI) of the β value for AUClast nearly fell within the theoretical limit (Figure 1I–L, Table S3).

The present data were references for designing an intracellular PK trial. Setting 1 dose can compare the PK feature between ZD2767P+CPG2 and ZD2767P+CPG2+US. At 100 or 200 μM ZD2767P, the drug concentration and “concentration × time” were within clinical Cmax and AUC, respectively; under 400 μM ZD2767P, “concentration × time” was above human AUC although the drug concentration was within Cmax.23 A low dose with short exposure duration was preferred since both the drug concentration and “concentration × time” accorded with human PK; i.e., results had better clinical relevancy. At least 3 doses were required for elucidating dose proportionality, and the dose ratio should be ≥2. The interval between CPG2 and insonation should allow all ZD2767P to be converted since CPG2 cannot enter cells: time (min) = amount of ZD2767P (mol)/[amount of CPG2 (U) × catalytic rate (mol·min–1·U–1)].4 An insonation level that can improve the CPG2 activity and membrane permeability with slight toxicity favored the emergence of the PK peculiarity.

Influx and efflux of drugs dynamically coexisted. A Ct vs t curve can be divided into the net-influx (before Cmax) and -efflux (after Cmax) phases (Figures S1–S3). At least 5 time points were set, with ≥3 time points (including Cmax) in the net-efflux phase for calculating MRT. Insonation immediately improved membrane permeability, which then decreased gradually until the preinsonation status.2 Thus, Cmax occurred shortly after insonation, as a guide to set time points in the net-influx phase. The present data can be references for exploring PK in other prodrug–enzyme systems.24−27

PK parameters can be used to optimize the therapeutic regimen. Cmax and AUClast were references to determine the ZD2767P dose. The treatment interval can be set based on MRTlast. Nearly equal modulation efficacies of ultrasound in sensitive and resistant cells indicated that a higher ZD2767P dose in resistant cancers was unnecessary, which can decrease toxicity.

In summary, the noncompartment model under extravascular mode was appropriate for evaluating intracellular PK of ZD2767D in the ZD2767P+CPG2+US system. PK parameters can be used to optimize the therapeutic regimen. Ultrasound equivalently modulated Cmax and AUClast in sensitive and resistant cells. The present data provided a reference for understanding the PK profiles in other prodrug–enzyme or prodrug–enzyme–ultrasound systems.

Experimental Procedures

Cells and Experimental Groups

Human lung cancer cell lines A549 and A549/DDP, and ovarian cancer cell lines SKOV3 and SKOV3/DDP were used (identified by short tandem repeats; Cell Bank, Type Culture Collect., Chin. Acad. Sci., Shanghai, China). A549/DDP and SKOV3/DDP can grow in 2 and 0.75 μg/mL cisplatin, respectively.3,7 Cells were cultured in RPMI 1640 medium (GIBCO, Beijing, China) enriched with 10% fetal bovine serum (Biol. Ind., Kibbutz Beit Haemek, Israel) and were transferred to cisplatin-free medium for 5 days before the experiments.

Cells received ZD2767P+CPG2 or ZD2767P+CPG2+US. ZD2767P and CPG2 were added to the cell suspension, followed by insonation (1.0 MHz, 10 W/cm2 for 20 s). Drugs were washed after 60 min. The dosage of ZD2767P (WuXi AppTec, Shanghai, China) was 100, 200, or 400 μM, and that of CPG2 (Chongqing Kerun Biomed. Pharm., Chongqing, China) was 1.2 U/mL. Insonation and/or CPG2 caused no cytotoxicity.3,7

Intracellular PK of ZD2767D

Intracellular ZD2767D was extracted at 0 (immediately after insonation), 5, 15, 30, 60, 90, and 120 min, and the concentration was measured by spectrophotometry.3 The drug level was normalized to the protein level. Ct vs t curves were analyzed using the software PKSolver (China Pharmaceutical University, Nanjing, China).28

Statistics

Data were processed with the software SAS (SAS Inst., Cary, NC, USA); analysis of variance was used, and the critical value was p < 0.05. Posthoc power (α = 0.05) was calculated to retrospectively evaluate the statistical power, using the software Stata (Stata Corp., College Station, TX, USA); the reference value was 0.80, the lowest priori power in designing a clinical trial.14,29 Ratios of geometric means of Cmax and AUClast (ZD2767P+CPG2+US / ZD2767P+CPG2) and the 90% CI were calculated to balance individual differences due to the resistance property and experimental manners, thereby comparing the PK modulation efficacy of ultrasound in sensitive and resistant cells.21

The dose proportionality was assessed with Cmax and AUClast, using the linear regression, quadratic linear regression, or power model. For the linear regression model (AUC = α + β × dose), the hypothesis was α = 0 and β > 0. For the quadratic linear regression model (AUC = α + β × dose + γ × dose2), the hypothesis was α = 0, γ = 0, and β > 0. For the power model (AUC = α × doseβ), the hypothesis was α > 0 and β = 1; proportionality was defined as the 90% CI of β value was within the theoretical limit (1+(ln[θL]/ln[r]), 1 + (ln[θU]/ln[r])); θL and θU were 0.8 and 1.25, respectively, and r was the ratio of the highest to lowest dose.22

Safety statement

No unexpected or unusually high safety hazards were encountered.

Glossary

Abbreviations

AUClast

area under the drug level vs time curve from zero to last measurable level

AUCinf

AUC from zero to infinity

Clast

last measurable level

Cmax

peak level

Ct vs t

level vs time

CI

confidence interval

Cl

clearance

CPG2

carboxypeptidase G2

EPR

enhanced permeability and retention effect

λz

slope of the terminal phase

MRTlast

mean residence time from zero to last measurable level

MRTinf

MRT from zero to infinity

PK

pharmacokinetics

t1/2

half-life

tlast

time of last measurable level

tmax

time of Cmax

US

ultrasound

Vz

volume of distribution

ZD2767P

4-[N,N-bis(2-iodoethyl)amino]phenoxycarbonyl-l-glutamic acid

ZD2767D

4-[N,N-bis(2-iodoethyl)amino]phenol

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.4c00071.

  • Ct vs t curves of ZD2767D at 100 μM ZD2767P; Ct vs t curves of ZD2767D at 200 μM ZD2767P; Ct vs t curves of ZD2767D at 400 μM ZD2767P; fitting Ct vs t curves of ZD2767D at 100 μM ZD2767P with PK models; t1/2 and MRT of ZD2767D at 100 μM ZD2767P in PK models; assessment of dose proportionality with Cmax and AUClast (PDF)

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Tianran Yu was funded with the Eyas Project from Chongqing Municipal Educational Commission.

The authors declare no competing financial interest.

Supplementary Material

ml4c00071_si_001.pdf (833.9KB, pdf)

References

  1. Sharma S. K.; Bagshawe K. D. Antibody directed enzyme prodrug therapy (ADEPT): trials and tribulations. Adv. Drug Delivery Rev. 2017, 118, 2–7. 10.1016/j.addr.2017.09.009. [DOI] [PubMed] [Google Scholar]
  2. Zhang Y.; Li J.; Yu T. Pharmacokinetic profiles of cancer sonochemotherapy. Expert Opin. Drug Delivery 2017, 14, 745–753. 10.1080/17425247.2016.1232248. [DOI] [PubMed] [Google Scholar]
  3. Liu Q.; Zhong X.; Zhang Y.; Li X.; Qian G.; Yu T. Ultrasound enhances ZD2767P–carboxypeptidase G2 against chemoresistant ovarian cancer cells by altering the intracellular pharmacokinetics of ZD2767D. Mol. Pharmaceutics 2020, 17, 1922–1932. 10.1021/acs.molpharmaceut.0c00008. [DOI] [PubMed] [Google Scholar]
  4. Yu T.; Li X. Development of ZD2767P–carboxypeptidase G2–ultrasound therapy against cisplatin-resistant cancer. Front. Oncol. 2023, 13, 1151613. 10.3389/fonc.2023.1151613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Romani A. M. P. Cisplatin in cancer treatment. Biochem. Pharmacol. 2022, 206, 115323. 10.1016/j.bcp.2022.115323. [DOI] [PubMed] [Google Scholar]
  6. Song M.; Cui M.; Liu K. Therapeutic strategies to overcome cisplatin resistance in ovarian cancer. Eur. J. Med. Chem. 2022, 232, 114205. 10.1016/j.ejmech.2022.114205. [DOI] [PubMed] [Google Scholar]
  7. Liu Q.; Li X.; Luo Y.; Wang H.; Zhang Y.; Yu T. Ultrasonically enhanced ZD2767P–carboxypeptidase G2 deactivates cisplatin-resistant human lung cancer cells. Oxid. Med. Cell Longev. 2022, 2022, 9191233. 10.1155/2022/9191233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Meng X.; Xu Y.; Lu Q.; Sun L.; An X.; Zhang J.; Chen J.; Gao Y.; Zhang Y.; Ning X. Ultrasound-responsive alkaline nanorobots for the treatment of lactic acidosis-mediated doxorubicin resistance. Nanoscale 2020, 12, 13801–13810. 10.1039/D0NR03726E. [DOI] [PubMed] [Google Scholar]
  9. Yildiz D.; Göstl R.; Herrmann A. Sonopharmacology: controlling pharmacotherapy and diagnosis by ultrasound-induced polymer mechanochemistry. Chem. Sci. 2022, 13, 13708. 10.1039/D2SC05196F. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Zhang Y.; Fowlkes B. Liposomes-based nanoplatform enlarges ultrasound-related diagnostic and therapeutic precision. Curr. Med. Chem. 2022, 29, 1331–1341. 10.2174/0929867328666210804092624. [DOI] [PubMed] [Google Scholar]
  11. Ren W. W.; Xu S. H.; Sun L. P.; Zhang K. Ultrasound-based drug delivery system. Curr. Med. Chem. 2022, 29, 1342–1351. 10.2174/0929867328666210617103905. [DOI] [PubMed] [Google Scholar]
  12. Kip B.; Tunc C. M.; Aydin O. Triple-combination therapy assisted with ultrasound-active gold nanoparticles and ultrasound therapy against 3D cisplatin-resistant ovarian cancer model. Ultrason. Sonochem. 2022, 82, 105903. 10.1016/j.ultsonch.2021.105903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Monks N. R.; Blakey D. C.; East S. J.; Dowell R. I.; Calvete J. A.; Curtin N. J.; Arris C. E.; Newell D. R. DNA interstrand cross-linking and TP53 status as determinants of tumour cell sensitivity in vitro to the antibody-directed enzyme prodrug therapy ZD2767. Eur. J. Cancer 2002, 38, 1543–1552. 10.1016/S0959-8049(02)00111-9. [DOI] [PubMed] [Google Scholar]
  14. Yu T.; Li X.. Clinical trials of high-intensity focused ultrasound for cancer: concerns arising from low post-hoc power. Curr. Med. Chem. 2024, 31 10.2174/0109298673281773240104142757. [DOI] [PubMed] [Google Scholar]
  15. Boroujerdi M., Ed. Pharmacokinetics and toxicokinetics ;CRC Press: Boca Raton, FL, 2015. [Google Scholar]
  16. Gabrielsson J.; Weiner D. Non-compartmental analysis. Methods Mol. Biol. 2012, 929, 377–389. 10.1007/978-1-62703-050-2_16. [DOI] [PubMed] [Google Scholar]
  17. Derieppe M.; Yudina A.; Lepetit-Coiffé M.; de Senneville B. D.; Bos C.; Moonen C. Real-time assessment of ultrasound-mediated drug delivery using fibered confocal fluorescence microscopy. Mol. Imaging Biol. 2013, 15, 3–11. 10.1007/s11307-012-0568-9. [DOI] [PubMed] [Google Scholar]
  18. Derieppe M.; de Senneville B. D.; Kuijf H.; Moonen C.; Bos C. Tracking of cell nuclei for assessment of in vitro uptake kinetics in ultrasound-mediated drug delivery using fibered confocal fluorescence microscopy. Mol. Imaging Biol. 2014, 16, 642–651. 10.1007/s11307-014-0726-3. [DOI] [PubMed] [Google Scholar]
  19. Lammertink B. H. A.; Deckers R.; Derieppe M.; De Cock I.; Lentacker I.; Storm G.; Moonen C. T. W.; Bos C. Dynamic fluorescence microscopy of cellular uptake of intercalating model drugs by ultrasound-activated microbubbles. Mol. Imaging Biol. 2017, 19, 683–693. 10.1007/s11307-016-1042-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Bahutair W. N.; Abuwatfa W. H.; Husseini G. A. Ultrasound triggering of liposomal nanodrugs for cancer therapy: a review. Nanomaterials 2022, 12, 3051. 10.3390/nano12173051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. U.S. Center for Drug Evaluation and Research . Voraxaze: clinical pharmacology and biopharmaceutics review. 2011. [Google Scholar]
  22. Hummel J.; McKendrick S.; Brindley C.; French R. Exploratory assessment of dose proportionality: review of current approaches and proposal for a practical criterion. Pharm. Stat. 2009, 8, 38–49. 10.1002/pst.326. [DOI] [PubMed] [Google Scholar]
  23. Mayer A.; Francis R. J.; Sharma S. K.; Tolner B.; Springer C. J.; Martin J.; Boxer G. M.; Bell J.; Green A. J.; Hartley J. A.; Cruickshank C.; Wren J.; Chester K. A.; Begent R. K. J. A phase I study of single administration of antibody-directed enzyme prodrug therapy with the recombinant anti-carcinoembryonic antigen antibody-enzyme fusion protein MFECP1 and a bis-iodo phenol mustard prodrug. Clin. Cancer Res. 2006, 12, 6509–6516. 10.1158/1078-0432.CCR-06-0769. [DOI] [PubMed] [Google Scholar]
  24. Sun I. C.; Yoon H. Y.; Lim D. K.; Kim K. Recent trends in in situ enzyme-activatable prodrugs for targeted cancer therapy. Bioconjugate Chem. 2020, 31, 1012–1024. 10.1021/acs.bioconjchem.0c00082. [DOI] [PubMed] [Google Scholar]
  25. Nervig C. S.; Hatch S. T.; Owen S. C. Complementation dependent enzyme prodrug therapy enables targeted activation of prodrug on HER2-positive cancer cells. ACS Med. Chem. Lett. 2022, 13, 1769–1775. 10.1021/acsmedchemlett.2c00394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ashoorzadeh A.; Mowday A. M.; Abbattista M. R.; Guise C. P.; Bull M. R.; Silva S.; Patterson A. V.; Smaill J. B. Design and biological evaluation of piperazine-bearing nitrobenzamide hypoxia/GDEPT prodrugs: the discovery of CP-506. ACS Med. Chem. Lett. 2023, 14, 1517–1523. 10.1021/acsmedchemlett.3c00321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Tantra T.; Singh Y.; Patekar R.; Kulkarni S.; Kumar P.; Thareja S. Phosphate prodrugs: an approach to improve the bioavailability of clinically approved drugs. Curr. Med. Chem. 2024, 31, 336–357. 10.2174/0929867330666230209094738. [DOI] [PubMed] [Google Scholar]
  28. Zhang Y.; Huo M.; Zhou J.; Xie S. PKSolver: an add-in program for pharmacokinetic and pharmacodynamic data analysis in Microsoft Excel. Comput. Methods Programs Biomed. 2010, 99, 306–314. 10.1016/j.cmpb.2010.01.007. [DOI] [PubMed] [Google Scholar]
  29. International Council for Harmonisation . Statistical principles for clinical trials. 1998. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ml4c00071_si_001.pdf (833.9KB, pdf)

Articles from ACS Medicinal Chemistry Letters are provided here courtesy of American Chemical Society

RESOURCES