( RESEARCH 49, 2844-2850. June 1, 1989] In Vivo Characterization of Combination Antitumor with Calcium Channel Blockers and ci5-Diamminedichloroplatinum(II)1

James M. Onoda,2 Kevin K. Nelson, John D. Taylor, and Kenneth V. Honn

Departments of Radiation [J. M. O., K. K. N., J. D. T., K. V. H.], Biological Sciences [J. D. T.], and Chemistry ¡K.V. H.], Wayne State University, Detroit, Michigan 48202; and the Cershenson Radiation Oncology Center [J. M. O., K. K. N., J. D. T., K. V. H.], Harper/Grace Hospitals, Detroit, Michigan 4820I

ABSTRACT dulin antagonists to enhance the antitumor actions of the more We have examined nifedipine, a dihydropyridine class calcium channel commonly prescribed organic or chemothera blocker, for ability to overcome m-diamminedichloroplatinum(II) (cis- peutic agents (3, 4). The ability of to reverse multi- platin) resistance in a murine tumor line variant, B16a-Pt, which we resistance or pleiotropic correlates with developed for resistance to . Nifedipine significantly enhanced the expression of a M, 170,000 glycoprotein in drug-resistant the antitumor actions of cisplatin against primary subcutaneous B16a-Pt tumor plasma membranes (5, 6). This glycoprotein is now tumors and their spontaneous pulmonary métastases.We have charac commonly referred to as the P-glycoprotein (7, 8) and is re terized, in vivo, the and dose-response interactions sponsible for the active efflux of many organic/natural product between nifedipine and cisplatin. We now report our studies designed to cytotoxic chemotherapeutic agents (9-11). The current hypoth compare, in vivo, the of nifedipine and other calcium active esis suggests that verapamil interacts with the P-glycoprotein compounds including: (a) structurally similar calcium channel blockers to block drug efflux (12, 13); and that its actions are independ (nimodipine, nicardipine) from the dihydropyridine class, (b) structurally ent of the classical slow-inward Ca2+ channel (14, 15). Calmo different calcium channel blockers from the benzothiazepine (diltiazem) and the phenylalkylamine (verapamil) classes, and (c) calmodulin antag dulin antagonists (16), CCBs of different chemical classes (13), onists (trifluoperazine and calmidazolium) for ability to enhance the and analogues of CCBs in a single class (17) often fail to work antitumor action of cisplatin. Nifedipine was included as the standard or as well as verapamil, suggesting that verapamil may interact reference compound. In these studies verapamil and diltiazem failed to with the P-glycoprotein at structurally definitive "binding sites" enhance the antitumor actions of cisplatin as did both calmodulin antag (18, 19). onists. Our findings suggest that nifedipine has a greater degree of In contrast to the abundant information available on MDR specificity for B16a-Pt cells than structurally different calcium channel less is known about tumor cell resistance to cisplatin, an inor blockers from other chemical classes (i.e., diltiazem and verapamil), or ganic cytotoxic agent. The P-glycoprotein may not be respon the two calmodulin antagonists u.c.. trifluoperazine and calmidazolium). sible for cisplatin-resistance, since cells exhibiting MDR resist We concluded that nifedipine interacts with specific target site(s) which are not accessible by verapamil, by diltiazem, or by the calmodulin ance usually remain sensitive to the cytotoxic effects of cisplatin antagonists. Surprisingly, the two dihydropyridine class calcium channel (20,21). blockers, nimodipine and nicardipine, also failed to enhance cisplatin's We recently reported (22, 23) that nifedipine, a dihydropyri antitumor actions despite the fact that their specificity and kinetics for dine class CCB, significantly enhances the cytotoxic antitumor binding to the dihydropyridine receptor component of the calcium channel and antimetastatic of effects of cisplatin against cisplatin-re favors them (nimodipine and nicardipine) over nifedipine. Therefore, we sistant primary tumors and their spontaneous pulmonary mé postulate that the synergism between cisplatin and nifedipine is inde pendent of the latter's effect on the voltage sensitive, slow inward calcium tastases. Our findings led us to test, in vivo, other calcium channel blockers of the dihydropyridine class, CCBs from dif channel. We suggest that cisplatin is enhanced by nifedipine's ferent chemical classes (i.e., benzothiazepine, phenylalkyla interaction with an as yet unidentified specific "target site," as opposed ) and calmodulin antagonists, to determine whether the to nonspecific interactions with the tumor cell plasma membrane or observed synergism between nifedipine and cisplatin was spe specific interactions with calmodulin or the P-glycoprotein (which is cific for nifedipine or was a general phenomenon exhibited by responsible for pleiotropic resistance). all classes of CCBs, by all dihydropyridine class CCBs or by calmodulin antagonists. INTRODUCTION Recently, novel chemotherapeutic agents and combination MATERIALS AND METHODS have resulted in promising new protocols for the treatment of recurrent and drug-resistant (1, Tumors 2). These regimens allow for an improved therapeutic efficacy but are often limited to a specific tumor type or to specific The B16 amelanotic (B16a) was originally obtained from cytotoxic agent(s). A significant advance in experimental ther the Mason Research Institute (Worchester, MA), Routine tumor trans apeutics involves the use of CCB3 (e.g., verapamil) and calmo- plantation was performed every 2 weeks into male, syngeneic host mice (C56BL/6 J; The Jackson Laboratory, Bar Harbor, ME). Mice used Received 9/20/88; revised 1/30/89: accepted 3/3/89. for transplantation were between 18 and 22 g (approximately 49 days The costs of publication of this article were defrayed in part by the payment old) and were housed under identical conditions of photoperiod, feeding of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. regime, temperature, etc. All animals were quarantined and acclimated 1This investigation was partially supported by a grant from the Meyer L. for a minimum of 10 days after receipt. Tumor brei (implanted into the Prentis Comprehensive Cancer Center of Metropolitan Detroit. Detroit. MI left axillary region) were used for routine transplantation. In order to 48201 [85-SC11: J. M. O.). and a grant from Harper Hospital |J. M. O., maintain metastatic phenotype, the B16a line was routinely restarted K. V. H.]. Detroit. Michigan, 48201. 2To whom requests for reprints should be addressed, at Department of from liquid N2 frozen stocks after eight to ten isotransplant generations Radiation Oncology, 431 Chemistry Building, Wayne State University, Detroit. in vivo. In addition, each line was recharacterized before use. Passages Michigan 48202. one and two after restarting each line from liquid N2 stocks were used ' The abbreviations and trivial name used are: CCB, calcium channel blockers; for characterization. Characterization included histology, transplanta- cisplatin. fM-diamminedichloroplatinum(ll): B16a. B16 amelanotic melanoma tumor line: B16a-Pt. cisplatin-resistant variant of the B16 amelanotic melanoma: bility, screen for murine tumor viruses, growth rate, metastatic pattern, 3LL, Lewis tumor line; MEM, minimal essential media; MDR. number of lung colonies formed in the experimental and multidrug resistance. spontaneous metastasis assay and doubling time in tissue culture. 2844

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for . COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN

Development of the Cisplatin-resistant Tumor Line B16a-Pt the antitumor effects of cisplatin with combination nifedipine , (c) to avoid death due to cisplatin, nifedipine and/or cisplatin/nifedipine The cisplatin-resistant B16a variant (B16a-Pt) line, was developed , and (d) to administer nifedipine and cisplatin to mice via the by a modification of the protocol of Schmid et al. (24). This was route most often used for human therapy, i.e., oral (nifedipine) and achieved by sequential treatment of primary subcutaneous tumors (ap intravenous (cisplatin). proximately 2 g) with two courses of cisplatin (4 mg/kg) on Days 14 Drug Dosage and Injection Schedule. We used drug dosage and and 21 posttumor brei implantation. On Day 33, primary tumors were injection protocols previously reported for combination nifedipine/ excised and adapted for growth in culture as described previously (22). cisplatin therapy against the B16a melanoma (22) and the Lewis lung The resulting primary cultures were treated with incremental (0.5-2.0 carcinoma (23). Nifedipine was administered orally (10 mg/kg; solubi- MM)exposure to cisplatin. Typically, exponentially growing cells were lized in 0.1 ml polyethylene glycol 400) 20 min prior to the i.v. (tail treated for one day with cisplatin, followed by removal of cisplatin, and vein) injection of cisplatin (4 mg/kg). were administered on Days readdition at the next highest dose. The cells which survived the in 14, 17, and 25 posttumor cell injection. Nifedipine, because it is vitro cisplatin treatment were reinjected s.c. into syngeneic mice. Mice absorbed from the gut, was administered 20 min prior to the injection bearing the resulting primary tumors were administered 4 mg/kg cis of cisplatin in order to have highest probability of maximum overlap platin as previously described (22, 23). After administration of cisplatin of peak plasma levels of the drugs. Drug(s) administration was first (4 mg/kg. Days 14 and 21 posttumor cell implantation), primary tumors performed on Day 14 posttumor cell injection, when the were excised and used to form primary cultures in vitro as described was both palpable and visible (approximately 0.5 g). Because cisplatin above. These cells were also treated with incremental (2.0 to 10 UM) administration leads to weight loss [an easily monitored sign of toxicity exposures to cisplatin. The surviving cisplatin-resistant cells were des in mice (29)], host animals were weighed daily following the initial ignated B16a-Pt. These cells proliferate in culture ( supple injection of cisplatin (Day 14). Days 17 and 25 posttumor injection mented MEM, with 10% fetal bovine serum), form subcutaneous tu were chosen for the second and third drug(s) administrations because mors and retain the parent B16a characteristic (25) of spontaneous weight loss had ceased and mouse weight had stabilized and/or was metastasis to the (22). Cisplatin-resistant cells were routinely increasing, indicating recovery from the prior drug administration. All exposed (in vitro) to a 10 ^M dose of cisplatin to maintain the drug- groups were terminated on Day 33 because the control group was resistant character of the line. moribund due to tumor burden. Primary tumors were removed, sized, For in vivo studies, monolayers of B16a-Pt cells were harvested with and weighed. Lungs were removed and fixed in Bouin's for at least 24 EDTA (2 mM), washed, and resuspended in MEM. Cells (IO5) were h. Macroscopic pulmonary tumor colonies were enumerated as previ injected s.c. into the left flank of host mice to form solid tumors. ously described (30). Isolation of Tumor Cells In Vivo Antitumor/Antimetastatic Studies

All of the studies presented utilized monodispersed cells obtained Primary s.c. B16a-Pt tumors were aseptically excised from host mice, from a primary subcutaneous tumor. Solid tumors were heterogeneous minced into 1-2-mm3 pieces and enzymatically dispersed with colla with respect to average tumor cell size and the proportion of either genase (28). Viable tumor cells were purified from nonviable tumor host cells, i.e., red cells, , etc., and of necrotic tumor cells and contaminating host cells by centrifugal elutriation (27). Mon cells. Thus, for our studies, cells were obtained by a modification of the odispersed B16a-Pt cells (3 x IO5cells) of high purity and viability were procedure described by Onoda et al. (26) and Ryan et al. (27). Subcu injected s.c. to form primary s.c. tumors. The criteria used to quantitate taneous tumors used for these isolations were of similar age (±2days) the antitumor effects of nifedipine/cisplatin were a reduction in primary and weight. Briefly, tumors were aseptically removed, minced, and tumor mass and spontaneous pulmonary métastases.Mice injected s.c. dispersed using sequential collagenase digestion (type CLS III; Wor- with B16a-Pt cells succumb to their tumor burden in approximately 33 thington Biochemical, Freehold, NJ). Centrifugal elutriation was used to 35 days. Therefore, studies (all groups) were sacrificed on Day 32. to separate viable from nonviable tumor cells and contaminating host Laboratory technicians who were blinded to treatment measured pri cells. The tumor cells used to initiate primary' s.c. tumors routinely mary tumor volume and enumerated tumor colonies. Primary tumors consisted of 100% monodispersed cells, of which >98% were tumor were measured using calipers and volume was calculated using the cells, of >95% viability, with no cellular debris (27, 28). equation: V = v/6 x LW2, where W is the center (of the tumor) measured perpendicular to /.. Lungs were excised and fixed for at least Cisplatin and Calcium Active Compounds 24 h in Bouin's and macroscopically visible tumor colonies were enu merated. Commercially available cisplatin was obtained from Bristol Labora tories. The compound [ci'i-diamminedichloroplatinum(II)] is available Statistical Analysis as a lyophilized powder containing 10 mg mannitol and 9 mg NaCl for each 1 mg of cisplatin. The preparation was suspended in sterile (glass) The distribution of in vivo tumor and métastasesdata is nonpara- distilled water at a concentration of 16 mg/ml [which is equivalent to metric, therefore we used the Kruskal-Wallis test to identify significant 4 mg/kg (2.7 mM) cisplatin for injection i.v. in a volume of 0.1 ml/ differences between groups. Groups demonstrating a statistically sig mouse]. nificant (P < 0.05) distribution were further analyzed by the Mann- The calcium channel blockers used were the commercial formula Whitney U test. tions available: nifedipine from Pharmaceuticals, Brooklyn, NY; verapamil from Knoll Pharmaceuticals, Whippany, NJ; diltiazem from RESULTS Marion Laboratories, Kansas City, MO; nicardipine from Syntex Inc., Palo Alto, CA; trifluoperazine from Sigma Chemical Co., St. Louis, We previously reported (22, 23) that nifedipine enhances the MO; and calmidazolium from Aldrich Chemical Co., Milwaukee, WI. antitumor actions of a's-diarnminedichloroplatinum(II) against Nimodipine was obtained through the courtesy of Dr. Alexander Scria- the inherently cisplatin-sensitive B16a and the inherently cis bine, Miles Institute for Preclinical , West Haven, CT. platin-resistant 3LL murine tumors. Subsequently, we devel oped a cisplatin-resistant variant (B16a-Pt) of the B16a tumor Initial Drug Dosage and Injection Schedule line and characterized in vivo the synergistic antitumor and Hypothesis. Our underlying hypothesis was that maximum minii antimetastatic actions of combined nifedipine/cisplatin ther pine/cisplatin antitumor effects would be achieved under conditions apy.4 where simultaneous maximum plasma levels (i.e., tumor exposure) of The data presented in Fig. 1 and 2 are representative of nifedipine and cisplatin were achieved. The initial parameters guiding studies characterizing the in vivoeffects of combined nifedipine/ our selection of drug dose levels and injection schedules were: (a) to obtain maximum antitumor effects with cisplatin alone, (b) to enhance 4 Manuscript submitted for publication. 2845

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN

B16a-Pt pine/cisplatin therapy results in a significant (P < 0.001) re duction of primary tumor volume when compared to the control group or the cisplatin treated group (P < 0.02). Similarly, the 125-, combination of nifedipine and cisplatin significantly reduced the incidence of pulmonary métastaseswhen compared to the control group (P < 0.001) or the group treated with cisplatin 100- alone (/>< 0.01). Half (6/12) of mice in the group administered both nifedipine and cisplatin were completely free of macro scopic pulmonary métastases. 75- CM Other calcium channel blockers were tested to determine the IO specificity of the interaction between nifedipine and cisplatin. uS 50- CO In addition to nifedipine, verapamil and diltiazem are two additional CCB commonly prescribed for cardiovascular-re in co lated therapy (31). They are also representative of two separate 25- and distinct classes of CCBs (i.e., diltiazem from the benzothi- azepine class, verapamil from the phenylalkylamine class), and have unique chemical structures from each other and from nifedipine (32). We suggest that the ability of nifedipine to C 10 NF 4 Gis 4 Gis ò^ + enhance the antitumor action of cisplatin is mediated by its 10NF interaction with specific cellular site(s) distinct from the voltage Fig. 1. Representative experiment demonstrating the efleet of nifedipine (10 sensitive calcium channel. Further, we suggest that the effects mg/kg), cisplatin (4 mg/kg) and nifedipine + cisplatin on primary B16a-Pt tumor observed with nifedipine may be unique to an element or volume. Nifedipine or cisplatin (alone) had no significant effect on tumor volume. However, treatment with nifedipine + cisplatin significantly reduced primary elements of its . Therefore, CCBs which are tumor volume (P < 0.001) when compared to the cisplatin alone treated group chemically distinct from nifedipine may not enhance the anti- or the control group. Bars, means ±SEM for each group, n = 12. The mean for the control was 2055 ±378-mm3 primary tumor volume. Inserts, median; range. tumor effects of cisplatin. We compared, in side-by-side studies, The study was repeated three times with comparable results. the efficacy of verapamil, diltiazem, and nifedipine as cisplatin enhancing agents. The data presented in Table I are represent Bl6a-Pt ative of these studies. The three CCBs were tested at an equal dose (10 mg/kg), soluhili/ed in the same solvent (Polyethylene- glycol 400), and administered orally (20 min prior to the tail vein injection of cisplatin). Treatment with cisplatin alone (4 0) 125-, oo mg/kg) had no effect on primary tumor volume or the incidence co of pulmonary métastases.Similarly, each of the channel block w ers, by itself, had no effect on primary tumor volume or the ~t—*CO 100- CD incidence of pulmonary métastases.In contrast, combination therapy with nifedipine significantly (P < 0.001) enhanced cisplatin's effects, whereas both verapamil and diltiazem failed 75- to enhance cisplatin's antitumor effects against primary tumors.

O Similarly, nifedipine significantly enhanced the antimetastatic j= 50- effects of cisplatin (as quantitated by a reduction in the inci U dence of pulmonary métastases),whereas both verapamil and Q_ diltiazem failed to enhance cisplatin's antimetastatic actions. 25- The pharmacokinetics for nifedipine, verapamil, and diltiazem are virtually identical (in mice, rats, and humans; Refs. 30, 33, O O 34), if careful pharmacokinetic studies and methodologies to à S C 10 NF 4CÎS 4 Gis Table 1 Enhancement of cisplatin's antitumor effects on the cisplatin-resistant 10NF variant B16a-Pt by calcium channel blockers Numbers represent median; range values, n = 12. Statistical reduction analyzed Fig. 2. Representative experiment demonstrating the effect of nifedipine (10 by the Kruschal-Wallis test. mg/kg), cisplatin (4 mg/kg) and nifedipine + cisplatin on the incidence of pulmonary métastases.Nifedipine or cisplatin (alone) had no significant effect on Mice the number of pulmonary métastases.However, treatment with nifedipine + tumor with cisplatin significantly reduced the incidence of pulmonary métastases(P< 0.001) GroupControl volume2260; métastases55; métastases100% when compared to the cisplatin alone treated group or the control group. Bars, 1178-5675 15-98 means ±SEM for each group, n = 12. The mean for the control was 38.8 ±6.2 Diltiazem (Dzm)° 1767; 1527-3591 40; 2 1-78 100% métastases.Inserts, median; range. The study was repeated three times with Verapamil (Ver)" 3002; 1521-6616 59; 43-85 100% comparable results. Nifedipine (NF)° 2137; 1288-4423 68; 47-83 100% Cisplatin' 1493; 804-3827 53; 38-70 100% Dzm + cisplatin 2293; 1288-3387 39; 16-92 100% cisplatin therapy against s.c. cisplatin-resistant tumors (B16a- Ver + cisplatin 1150; 827-3315 37; 4-97 100% 462; 78-954cPulmonary0-,0-ìì"% Pt) and their spontaneous pulmonary métastases.The drug(s) NF + cisplatinPrimary 42% dose levels, sequence, , and days of " 10 mg/kg. administration were those previously discussed (see "Meth ' 4 mg/kg. ' Nifedipine + cisplatin significantly reduces primary tumor volume when ods"). Nifedipine (10 mg/kg, orally) alone, or cisplatin (4 mg/ compared to the control group (/' < 0.001) or the groups treated with nifedipine alone (P < 0.001 ) or cisplatin alone (P < 0.001 ). kg, i.v.) alone, had no effect on primary tumor volume (Fig. 1) ' Nifedipine + cisplatin significantly reduces pulmonary métastaseswhen or on the incidence of pulmonary métastases(Fig. 2) when compared to the control group (P < 0.001) or the groups treated with nifedipine compared to the control group. In contrast, combined nifedi- alone (P < 0.001) or cisplatin alone (P < 0.001). 2846

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN B16a-Pt Table 2 Enhancement of cisplatin 's antitumor effects on cisplatin-resistant variant BI6a-Pt by the dihydropyridines 160 - Numbers represent median; range values, n = 12. Statistical reduction analyzed by the Kruschal-Wallis test. 140 - mice tumor with o 120 GroupControl volume4007; métastases37;métastases100% >, 100 - 2532-5962 22-75 ra Nimodipine (NM)° c 5325; 2389-6355 35; 22-71 100% o Nicardipine (NC)° 80 • 4546; 1100-7889 43; 23-60 100% Nifedipine (NF)° 3541;2262-9425 37; 31-42 100% Cisplatin* O. 2655:2572-3591 35; 23-59 100% 60 NM + cisplatin 3913:2806-7238 43; 27-64 100% 3882; 1923-5221 31; 26-52 I 40- NC + cisplatin 100% S NF + cisplatinPrimary 947; 385-3053'Pulmonary7; 0-37'% 75% 20 - "10 mg/kg. 5? *4 mg/kg. ' Nifedipine + cisplatin significantly reduces primary tumor volume when O 20 60 80 100 compared to the control group (P < 0.001) or the groups treated with nifedipine alone (P < 0.001) or cisplatin alone (P < 0.001). ''Nifedipine + cisplatin significantly reduces pulmonary métastaseswhen [mg/kg] compared to the control group (P < 0.001) or the groups treated with nifedipine Fig. 3. Representative experiment demonstrating the effect of varying doses alone (P < 0.001 ) or cisplatin alone (P < 0.001 ). of nifedipine (O), diltiazem (•),or verapamil (D) administered in combination with 4 mg/kg cisplatin on the incidence of B16a-Pt pulmonary métastases.A dose-response effect was clearly obtained with increasing levels of nifedipine. Significant reduction in the incidence of pulmonary métastases(/' < 0.001) B16a-Pt relative to the group treated with cisplatin alone (data point 0.0 mg/kg nifedipine), was achieved with doses of 2.5, 5.0, 10.0, and 25.0 mg/kg (+4 mg/kg cisplatin). Nifedipine at 25 mg/kg and 50 mg/kg were fatally toxic to 25% and 100% animals, respectively. Diltiazem and verapamil failed to enhance the antimeta- static effects of cisplatin at all doses tested (10-100 mg/kg). The slight reduction

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN B16 a - Pt classes of organic or natural product cytotoxic chemotherapeu tic agents (e.g., , ; Ref. 125-, 9, 10). Verapamil decreases the efflux of chemotherapeutic 10 drugs from drug-resistant tumor cells via its interaction with sin the P-glycoprotein (12,13,18,19), thus, inhibiting its function. a 100- o> This results in an increased intracellular accumulation of the cytotoxic agent (3) and a concomitant decrease in the apparent concentration of the antitumor agent necessary to achieve LD5o a 75- kill rates. In addition to its interaction with the P-glycoprotein, O E verapamil may enhance drug cytotoxicity via other mechanisms. a Q. 50- Harker et al. (39) suggests that verapamil, by interaction with II, enhances the effects of and other agents which are cytotoxic due to their DNA intercalation oC 25- properties. They suggest that verapamil acts by inhibiting repair U of DNA strand breaks (by topoisomerase II) or by permitting •*- O increased formation of DNA strand breaks. The success of S« verapamil (in combination chemotherapy) has stimulated the Control Gis CZM TFP NF screening of CCBs from other chemical classes for their ability CÃŽs Cls Cls to reverse MDR (3, 4). Diltiazem and several dihydropyridine Fig. S. Representative experiment demonstrating the effect of calmodulin antagonists (calmidazolium and trifluoperazine; 10 mg/kg) in combination with class CCBs (including nifedipine) have, in many studies, dem 4 mg/kg cisplatin on the incidence of B16a-Pt pulmonary métastases.The group onstrated ability equal to that of verapamil for reversing MDR treated with nifedipine (10 mg/kg) + cisplatin (4 mg/kg) significantly (P< 0.001) (5). However, in other studies, diltiazem and dihydropyridine decreased the incidence of pulmonary métastaseswhen compared to the control group or the groups treated with cisplatin alone. Groups treated with calmida CCBs were, at best, 10- to 100-fold less potent than verapamil zolium + cisplatin or trifluoperazine + cisplatin failed to significantly reduce the (17). This suggests that the interaction between CCBs with incidence of pulmonary métastaseswhen compared to the control group or the group treated with cisplatin alone. Bars, means ±SEM; n = 12. Inserts, median; cellular elements responsible for drug resistance (a) may be range. [Not shown: calmidazolium alone (10 mg/kg) and trifluoperazine alone very specific, (b) appear to be unique to each species of CCB, (10 mg/kg) caused no significant reduction on the incidence of pulmonary and (c) are independent of their binding affinity and classical métastases,95%and 102% of control, respectively.] The study was repeated three times with comparable results. function as inhibitors of the voltage-sensitive calcium channel. In contrast to the identified mechanism for MDR (i.e., the action of the P-glycoprotein), less is known about the mecha spontaneous pulmonary métastases.The drug(s) dose levels, nism^) responsible for cisplatin resistance. It is now generally sequence, route of administration, and days of administration, acknowledged that the P-glycoprotein responsible for MDR is were described previously (see "Methods"). Nifedipine was not responsible for cisplatin resistance (20, 21). In addition, a included in these studies as a positive control. The combination review of the literature suggests a lack of consensus as to the of each calmodulin antagonist (10 mg/kg) with cisplatin (4 mg/ identity of a single mechanism (analogous to the P-glycopro kg) failed to enhance the effect of cisplatin on tumor volume tein) responsible for cisplatin-resistance. An earlier report sug (Fig. 4) or the incidence of pulmonary métastases(Fig. 5). As gests few if any differences in cisplatin uptake or efflux among anticipated, the combination of nifedipine (10 mg/kg) plus cisplatin-sensitive and cisplatin-resistant human var cisplatin (4 mg/kg) significantly reduced primary tumor volume iants (40). Thus, the principal mechanism of tumor cell resist (P < 0.001) and the incidence of pulmonary métastases(P < ance to cisplatin may not be increased drug efflux (41, 42). 0.001). In addition, neither trifluoperazine (10 mg/kg, oral) However, Richon et al. (43) reported small but significant alone, calmidazolium (10 mg/kg, oral) alone, or cisplatin (4 differences in cellular accumulation of cisplatin between cispla mg/kg, i.v.) alone, were effective against B16a-Pt tumor volume tin-sensitive and -resistant tumor variants. Others suggest that (Fig. 4) or on the incidence of pulmonary métastases(Fig. 5) an enhanced ability to repair cisplatin-induced DNA-DNA (see figure legends). cross-linking is the principal mechanism of resistance (4, 44). Recent studies describe cisplatin resistance as (a) a function of recessive traits causing reduced cell uptake of cisplatin and DISCUSSION increased intracellular levels of glutathione (43), (b) related to The application of novel agents for use in combination ther DNA-protective -binding (; 45), and (c) mediated by Ca2+ transport across tumor cell endo- apies with cytotoxic drugs is necessitated by the dilemma of drug-resistant tumors. Principal among these agents are cal plasmic reticulum (46). Therefore, at the present time, the cium channel blockers which enhance the antitumor actions of mechanism(s) of cisplatin resistance may be described as mul- certain classes of the organic or natural product cytotoxic tifactorial, in as much as the available evidence clearly fails to chemotherapeutic agents. Tsuruo et al., (38) first reported that identify any singular mechanism wholly responsible for tumor in vitro and in vivo the phenylalkylamine class calcium channel cell resistance. blocker, verapamil, overcame or reversed resistance Previous attempts to enhance the antitumor actions of cis in the murine P388 leukemia tumor line. Tsuruo's subsequent platin involved changes in the treatment protocols, including reports (3, 16, 17) and those of other investigators have (4,15) scheduling, mode of administration and dosage (47, 48). In clearly established the ability of verapamil and other calcium addition, combination therapy with other cytotoxic chemother regulatory compounds, (e.g., the calmodulin antagonist, trifluo apeutic agents such as , and/or 5-flurou- perazine) to enhance the cytotoxic effect of anticancer agents racil are now used routinely (1, 2). Few investigators have against drug-resistant tumors. examined the ability of noncytotoxic agents to enhance the Pleiotropic or multidrug resistance is mediated by the M, antineoplastic effects of cisplatin, perhaps because the mecha 170,000, transmembrane P-glycoprotein which extrudes certain nism of cisplatin resistance remains poorly understood. Those 2848

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN agents which were examined (i.e., misonidazole and verapamil) cokinetic or differences among the calcium chan produced negative results (49, 50). nel blockers and calmodulin antagonists are a less likely expla We reported previously that nifedipine enhances the cyto- nation for our findings. toxic antitumor actions of cisplatin /// vitro and in vivo against The need to improve the efficacy of conventional cytotoxic cisplatin-sensitive and cisplatin-resistant murine tumors and chemotherapeutic agents is emphasized by the research effort their spontaneous pulmonary métastases(22, 23). Now we devoted to the development of combination chemotherapeutics report the (a) in vivo characterization and comparison of CCB and novel protocols. The primary focus of this regimen is on from different chemical classes (i.e., phenylalkylamine class, the treatment of recurrent or drug-resistant in a manner benzothiazepine class, and dihydropyridine class), (A) compar which limits toxicity to the host. Our report demonstrating the ison of three agents (nimodipine, nicardipine, and nifedipine) ability of the noncytotoxic agent, nifedipine, to synergistically from the same class of CCBs, and (c) analysis of two caimodulili enhance the antitumor actions of cisplatin against cisplatin- antagonists (i.e., trifluoperazine and calmidazolium) for their resistant murine tumors suggests that this combination protocol ability to enhance the cytotoxic antitumor effects of cisplatin may provide therapeutic benefits in the treatment of human in vivo against a murine tumor line (B16a-Pt) with developed and recurrent disease. resistance to cisplatin. Our comparative studies reveal no significant synergism be REFERENCES tween diltiazem and cisplatin or verapamil and cisplatin even at suprapharmacological doses when compared to drug-free 1. Einhorn, L. H., and Donohue, J. C/s-diamminedichloroplatinum, vinblastine and bleomycin combination therapy in disseminated . Ann. control group animals or groups receiving verapamil, diltiazem, Int. Med., 87: 293-298, 1977. or cisplatin alone. In addition, calmodulin antagonists also 2. Randolph. V. L., Vallejo, A., Spiro, R. H., Shah. J., Strong, E., Huvos, A., failed to enhance cisplatin's antitumor actions. Our findings and Wittes, R. Combination therapy of advanced . Cancer (Phila.). 41: 460-467, 1978. suggest that nifedipine has a greater degree of specificity for 3. Tsuruo, T. Reversal of acquired resistance to Vinca alkaloids and anthracy- B16a-Pt cells than representative agents from two other classes cline antibiotics. Cancer Treat. Rep., 67: 889-894, 1983. 4. Helson, L. enhancement of anticancer drug cyto- of CCBs or calmodulin antagonists. Thus, we conclude that toxicity—a review. Cancer , /: 353-361, 1984. nifedipine may interact with specific cellular target site(s) with 5. Tsuruo, T., Kawabata, H., Nagumo, N., lida, H., Kitatani, V. Tsukagoshi, S., and Sakurai, Y. Potentiation of antitumor agents by calcium channel greater intrinsic potency than verapamil, diltiazem, or the cal blockers with special reference to cross-resistance patterns. Cancer Chemo- modulin antagonists. Moreover, nimodipine and nicardipine ther. Pharmacol., /5: 16-19, 1985. are ineffective despite that fact that their kinetics for binding 6. Cano-Gauci, D. F., and Riordan, J. R. Action of calcium antagonists on multidrug resistant cells. Biochem. Pharmacol., 36: 2115-2123, 1987. to the dihydropyridine receptor component of the calcium 7. Kartner, N., Evernden-Porelle, D., Bradley, G., and Ling, V. Detection of P- channel favors them (nimodipine and nicardipine) over nifedi glycoprotein in multidrug-resistant cell lines by monoclonal . Na pine (36, 37). Therefore, we postulate that the synergism be ture, (Lond.), 316: 820-823, 1985. tween cisplatin and nifedipine is independent of the latter's 8. Juliano, R. L., and Ling, V. A surface glycoprotein modulating drug perme ability in Chinese hamster ovary cell mutants. Biochim. Biophys. Acta, 455: effect on the voltage-sensitive, slow inward calcium channel. 152-162,1982. 9. Dane, K. Active outward transport of daunomycin in resistant Ehrlich ascites Nevertheless, we cannot discount the possibility that the ob tumor cells. Biochim. Biophys. Acta, 323:466-483, 1973. served synergism between nifedipine and cisplatin is the result 10. Skovsgaard, T. Mechanisms of resistance to in Ehrlich ascites of nifedipine's ability to alter intracellular levels of calcium via tumor cells. Cancer Res., 38: 1785-1791, 1978. mechanisms independent of the voltage-sensitive calcium chan 11. Gros, P., Neriah, Y. B., Croop, J. M., and Houseman, D. E. Isolation and expression of a complementary DNA that confers multidrug resistance. nel. For example, calcium channel blockers are known to inhibit Nature (Lond.), 323: 728-731, 1986. Ca2"1"influxinto although platelets lack voltage-sensi 12. Cornwell, M. M., Safa, A. R., Felsted, R. L., Gottesman, M. M., and Pastan, I. Membrane vesicles from multidrug-resistant human cancer cells contain a tive calcium channels (51, 52). specific 150- to 170-kDa detected by photoaffmity labeling. Proc. One may question the significance of pharmacokinetic differ Nati. Acad. Sci. USA, 83:3847-3850, 1986. ences or differences in the bioavailability between nifedipine 13. Safa, A. R., Glover, C. J., Meyers. M. B., Biedler, J. L., and Felsted, R. L. Vinblastine photoaffmity labeling of a high molecular weight surface mem and the other CCBs as an alternative explanation which could brane glycoprotein specific for multidrug-resistant cells. J. Biol. < IK-HI..261: account for the unique superiority of nifedipine to enhance the 6137-6141. 1986. 14. Ramu, A., Spanier, R., Rahamimoff, H., and Fuks, Z. Restoration ofdoxo- antimetastatic and antitumor actions of cisplatin. However, as described in "Methods," our drug administration protocol min rubicin-resistant P388 murine leukemia cells. Br. J. Cancer, 50: 501-507, 1984. imizes differences in the relative bioavailability of the channel 15. Kessel. D., and Wilberding, C. Anthracycline resistance in P388 murine leukemia and its circumvention by calcium antagonists. Cancer Res., 45: blockers and calmodulin antagonists. Most important to this 1687-1691, 1985. consideration is the recognition that our protocol utilizes a 16. Tsuruo, T., lida, H., Tsukagoshi, S., and Sakurai, Y. Increased accumulation of vincristine and Adriamycin in drug-resistant P388 tumor cells following solubilized powdered form of the drugs for oral administration, incubation with calcium antagonists and calmodulin inhibitors. Cancer Res., thereby facilitating very rapid absorption from the gut (as «.•4730-4733,1982. opposed to use of slower acting capsules for drug administra 17. Tsuruo. T.. lida. H.. Nojiri, M., Tsukagoshi, S., and Sakurai. Y. Circumven tion of vincristine and adriamycin resistance in litro and in vivo by calcium tion). Furthermore, most studies which examine drug levels influx blockers. Cancer Res., 43: 2905-2910, 1983. and determine times for onset of drug action (in humans) 18. Beck, W. T. The cell biology of . Biochem. Pharma measure total drug levels and gross physiological drug effects col., 36: 2879-2887, 1987. 19. Inaba, M.. and Maruyama. E. Reversal of resistance to vincristine in P388 (e.g., P/R intervals; Refs. 31, 34). More specific measurement leukemia by various polycyclic clinical drugs, with a special emphasis on of free or bioactive drug (i.e., unbound to serum albumins; Refs. quinacrine. Cancer Res., 48: 2064-2067, 1988. 20. Seeber, S., Osieka, R., Schmidt, C. G., Acherrath, W., and Crooke, S. T. In 53-57) after oral administration yield onset times of 10 to 30 vivo resistance towards , , and r/'j-diamminedichlo- min (55-57) and peak drug effects of 20-30 min to 2 h, ropl.itmum. Cancer Res.. 42:4719-4725, 1982. 21. Supino. R., Prosperi, E., Formelli, F., Mariani, M., and Parmiani, G. especially if biochemical effects (54), as opposed to gross phys Characterization of a doxorubicin-resistant murine melanoma line: studies iological effects, are examined. Thus, our drug administration on cross-resistance and its circumvention. Br. J. Cancer, 54: 33-42, 1986. protocol optimizes the probability of maximum overlap of peak 22. Onoda, J. M., Jacobs, J. R. Taylor, J. D., Sloane, B. F., and Honn, K. V. Cisplatin and nifedipine: synergistic cytotoxicity against murine solid tumors plasma levels of drugs. Although we cannot completely exclude and their métastases.Cancer Lett.. 30: 181-188. 1986. this explanation at the present time, we believe that pharma 23. Onoda, J. M., Nelson, K. K., Taylor, J. D., and Honn, K. V. Cisplatin and 2849

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. COMBINATION CHEMOTHERAPY WITH CALCIUM ANTAGONISTS AND CISPLATIN

nifedipine: synergistic antitumor effects against an inherently cisplatin-re- systems in K562 cells sensitive and resistant to ds-diamminedichloro- sistant tumor. Cancer Lett., 40: 39-47, 1988. (II). Cancer Res., 46: 3445-3448, 1986. 24. Schmid, F. A., Otter, J. M., and Stock, C. C. Resistance patterns of walker 41. Fojo, A., Akiyama, S., Gottesman, M. M., and Pastan, I. Reduced drug carcinosarcoma 256 and other rodent tumors to and L- accumulation in multiple drug resistant human KB carcinoma cell lines. phenylalanine mustard. Cancer Res., 40: 830-833, 1980. Cancer Res., 45: 3002-3007, 1985. 25. Honn, K. V., Onoda, J. M., Pampalona, K., Battaglia, M., Neagos, G., 42. I rum, R. J., Cusik, P. S., Wilson, J. M., and Marinus, M. G. Mismatch Taylor, J. I).. Diglio, C. A., and Sloane, B. F. Inhibition of dihydropyridine repair of cu-diamminedichloroplatinum(II)-induced DNA damage. Mol. class calcium channel blockers of tumor cell--endothelial cell inter Pharmacol., 2«:51-55, 1985. actions in vitro and metastasis in vivo. Biochem. Pharmacol., 34: 235-241, 43. Richon, V. M., Schulte, N., and Eastman, A. Multiple mechanisms of 1985. resistance to cis-diamminedichloroplatinum(II) in murine leukemia LI210 26. Onoda, J. M., Nelson, K. K., Grossi, I. M., Umbarger, L. A., Taylor, J. D., cells. Cancer Res., 47: 2056-2061, 1987. and Honn, K. V. Separation of high and low metastatic subpopulations from 44. Behrens, B. C., Hamilton, T. C., Masuda, H., Gratzinger, K. R., Whang- solid tumors by centrifugal elutriation. Proc. Soc. Exp. Biol. Med., 187:250- Peng, J., Louie, K. G., Knutsen, T., McKoy, W. M., Young, R. C., and 255, 1988. Ozols, R. F. Characterization of a cu-diamminedichloroplatinum(II)-resist- 27. Ryan, K.. Crissman, J. D., Honn, K. V., and Sloane, B. F. Cathepsin B-like ant human ovarian line and its use in evaluation of platinum activity in viable tumor cells isolated from rodent tumors. Cancer Res., 35: analogues. Cancer Res., 47:414-418, 1987. 3636-3641, 1985. 45. Metcalfe, S. A., Cain, K., and Hill, B. T. Possible mechanism for differences 28. Honn, K. V., Cicone, B., and Skoff, A. Prostacyclin: a potent antimetastatic in sensitivity to m-platinuni in human tumor cell lines. Cancer agent. Science (Wash. DC), 212:1270-1272, 1981. Lett., 31:163-168, 1986. 29. Chahinian, A. P., Norton, L., Holland, J. F., Szrajer, L., and Hart, R. D. 46. Vassilev, P. M., Kanazirska, M. P., Charamella, L. J., Dimitrov, N. V., and Experimental and clinical activity of and c/i-diamminedichlo- Tien, H. T. Changes in calcium channel activity in membranes from civ- roplatinum in malignant . Cancer Res., 44:1688-1692, 1984. diamminedichloroplatinum(ll)-resistant and -sensitive L1210 cells. Cancer 30. Onoda, J. M., Sloane, B. F., Taylor, J. D., and Honn, K. V. Calcium channel Res., 47:519-522, 1987. blockers: Inhibitors of tumor cell-platelet-endothelial cell interactions. In: K. 47. Amer, M. H., Al-Sarraf, M., and Vaitkevicius, V. Factors that effect response V. Honn and B. F. Sloan (eds.), Hemostatic Mechanisms of Metastasis, pp. to chemotherapy and survival of patients with advanced head and neck cancer. 244-258. The Hague: Martinus-Nijhoff, 1984. Cancer (Phila.), 43:1202-1206, 1979. 31. Schwartz, D. J., Wasserstrom, J. A„andFozzard, H.A. Therapeutic uses of 48. Robins, A. B., and Leach, M. O. Pharmacokinetics of therapeutic doses of calcium-blocking agents: verapamil, nifedipine and diltiazem. Comprehensive isotopically labeled platinum antitumor agents in the mouse and rat. Cancer Ther., 7:25-33, 1981. Treat. Rep., 67: 245-252, 1983. 32. Janis, R. A., Silver, P. J., and Triggle, D. J. Drug action and cellular calcium 49. Clement, J. J., Gorman, M. S., Wodinski, I., Catane, R., and Johnson, R. regulation. Adv. Drug Res. 16: 309-591, 1987. V. Enhancement of antitumor activity of alkylating agents by the radiation sensitizer misonidazole. Cancer Res., 40:4165-4172, 1980. 33. Malo, P., Wasserman, M. A., and Griffin, R. L. The effects of nifedipine and verapamil on antigen-induced broncoconstriction in dogs. Eur. J. Pharmacol., 50. Mizuno, S., and Ishida, A. Potentiation of bleomycin cytotoxicity by mem brane-interacting drugs and increased calcium ions. Biochem. Biophys. Res. 92:69-75, 1983. Commun., 707:1021-1030,1982. 34. Conti, C. R. Calcium channel blockers: rationale for their clinical use. J. 51. Motulsky, H. J., Snavely, M. D., Hughes, R. J., and Insel, P. A. Interaction Florida Med. Assoc., 68: 883-886, 1981. of verapamil and other calcium channel blockers with «-1and a-2 adrenecyin 35. Ikeda, II., Nakano, G., Nagashima, K., Sakamoto, K., Harasawa, N., Kita- receptors. Circ. Res., 52: 226-231, 1983. mura, T., Nakamura, T., and Nagamachi, Y. Verapamil enhancement of 52. Onoda, J. M., Sloane, B. F., and Honn, K. V. Antithrombogenic effects of antitumor effects of rij-diamminedichloroplatinum(II) in nude mouse-grown calcium channel blockers: synergism with prostacyclin and thromboxane human neuroblastoma. Cancer Res., 47: 231-234, 1987. synthase inhibitors. Thromb. Res., 34: 367-378, 1984. 36. Epstein, P. M., Fiss, K., Hachisu, R., and Andrenyak, D. M. Interaction of 53. McAllister, R. G., and Kirsten, E. B. The pharmacology of verapamil. IV. calcium antagonists with cyclic AMP phosphodiesterases and calmodulin. Kinetic and dynamic effects after single intravenous and oral doses. Clin. Biochem. Biophys. Res. Commun., 105: 1142-1149, 1982. 37. Janis, R. A., and Triggle, D. J. New developments in (.'a1' channel antago Pharmacol. Ther., 31: 418-426, 1982. 54. McAllister, R. G., Hamann, S. R., and Blouin, R. A. Pharmacokinetics of nists. J. Med. Chem., 26: 775-785, 1983. calcium-entry blockers. Am. J. Cardio!., 55: 30B-40B, 1985. 38. Tsuruo, T., lida, II.. Tsukagoshi, S., and Sakurai, Y. Overcoming of vincris- 55. Hamann, S. R., Todd, G. D., and McAllister, R. G. The pharmacology of tine resistance in P388 leukemia in vivo and in vitro through enhanced verapamil. Pharmacology, 27: 1-8, 1983. cytoxicity of vineristi ne and vinblastine by verapamil. Cancer Res., 41:1967- 56. Eichelbaum, M., Somogyi, A., von Unruh, G. E., and Dengler, H. J. Simul 1972, 1981. taneous determination of the intravenous and oral pharmacokinetic param 39. Harker, W. G., Bauer, D., Etiz, B. B., Newman, R. A., and Sikic, B. I. eters of D.I-verapamil using stable isotope labelled verapamil. Eur. J. Clin. Verapamil-mediated sensitization of doxorubicin-selected pleiotropic resist Pharmacol., 19: 133-137, 1981. ance in human cells: selectivity for drugs which produce DNA 57. McAllister, R. G., Tan, T. G., and Bourne, D. W. A. GLC assay of verapamil scission. Cancer Res., 46: 2369-2373,1986. in plasma: identification of fluorescent metabolites after oral drug adminis 40. Shionoya, A., Lu, Y., and Scanion, K. J. Properties of amino acid transport tration. J. Pharm. Sci., 6«:574-578, 1979.

2850

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research. In Vivo Characterization of Combination Antitumor Chemotherapy with Calcium Channel Blockers and cis -Diamminedichloroplatinum(II)

James M. Onoda, Kevin K. Nelson, John D. Taylor, et al.

Cancer Res 1989;49:2844-2850.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/49/11/2844

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/49/11/2844. Click on "Request Permissions" which will take you to the Copyright Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1989 American Association for Cancer Research.