Phase I trial of retinoic acid and cis-platinum for advanced squamous cell cancer of the head and neck based on experimental evidence of drug synergism.

OBJECTIVE: Cis-platinum and 13-cis-retinoic acid have received much attention in the treatment of head and neck squamous cell cancer. Even though they have different mechanisms of action, little information is available on their interaction. This paper reviews experimental evidence for retinoic acid-cis-platinum synergy and presents toxicity data from patients with stage IV head and neck squamous cell cancer participating in a phase I trial combining 13-cis-retinoic acid and cis-platinum. METHODS: Patients were given 13-cis-retinoic acid orally daily for 7 days before and daily during high-dose (150 mg/m2 per week for 4 weeks) intraarterial cis-platinum treatment with concurrent radiation. Toxicity was scored with use of the cancer and leukemia group B scale. RESULTS: In the phase I clinical trial, 15 patients were treated to determine a maximum tolerated dosage for 13-cis-retinoic acid of 20 mg/day. Grade 4 hematologic toxicity was dose limiting in 3 of 8 patients treated with 40 mg/day and in 1 patient treated with 60 mg/day. There were no deaths caused by toxicity; 12 of the 15 patients received all four weekly doses and the remaining 3 received three doses. Of 10 patients with fully evaluable data, all achieved a complete response at the primary site and 9 had a complete response in the neck. One patient had persistent neck disease after chemoradiation, and this tumor was removed with neck dissection. CONCLUSIONS: 13-Cis-retinoic acid and cis-platinum are strongly synergistic against head and neck squamous cell cancer in vitro. Pretreatment with retinoic acid results in stronger synergy than concurrent drug exposure alone. Preliminary clinical experience with combined retinoic acid and cis-platinum in a design that parallels the in vitro study indicates that toxicity is acceptable with 13-cis-retinoic acid dosages of 20 mg/day in a high-dose-intensity intraarterial chemoradiation regimen.

Phosphoinositide signalling in human neuroblastoma cells: biphasic effect of Li+ on the level of the inositolphosphate second messengers.

Lithium has a biphasic effect of the agonist-dependent accumulation of Ins(1,4,5)P3 in human neuroblastoma SH-SY5Y cells. These effects consist of a transient reduction, followed by a long-lasting increase in Ins(1,4,5)P3 as compared to controls. The Li+ effects are dose dependent, and were observed at concentrations used in the treatment of bipolar disorders, and thus may have therapeutic implications. The mechanism of the Li+ effect on Ins(1,4,5)P3 accumulation requires further investigation. The transient reduction of Ins(1,4,5)P3 was observed under conditions where Li+ causes only a moderate increase in the inositol mono- and bi-phosphates. Supplementation with exogenous inositol had no effect on the level of Ins(1,4,5)P3, indicating that the mechanism of the Li(+)-dependent reduction of Ins(1,4,5)P3 is not due to inositol depletion. Li+ did not interfere with degradation of Ins(1,4,5)P3 after receptor-blockage with atropine, suggesting that Li+ has no direct effect on the Ins(1,4,5)P3 metabolizing enzymes. A direct effect of Li+ on the phospholipase C is also unlikely. Entry of Ca2+ into the cells is an important factor, which affects agonist-stimulated accumulation of Ins(1,4,5)P3, as well as absolute values of Li(+)-dependent increase in Ins(1,4,5)P3; however, it is not essential for the manifestation of Li+ effects. Our results also show that manifestation of Li+ effects in human neuroblastoma cells requires the stimulation of muscarinic receptors and activation of PLCs, PKCs, and/or that other staurosporine/H-7/GF 109203X-sensitive protein kinases are involved in the regulation of Ins(1,4,5)P3 during the plateau phase of ACh-stimulation. We also suggest an important role for these enzymes in the Li(+)-dependent elevation of Ins(1,4,5)P3.

Time-dependent effects of lithium on the agonist-stimulated accumulation of second messenger inositol 1,4,5-trisphosphate in SH-SY5Y human neuroblastoma cells.

In order to approach the molecular mechanism of Li+'s mood-stabilizing action, the effect of Li+ (LiCl) on inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] mass was investigated in human neuroblastoma SH-SY5Y cells, which express muscarinic M3 receptors, coupled to PtdIns hydrolysis. Stimulation of these cells, with the cholinergic agonist acetylcholine, resulted in a rapid and transient increase in Ins(1,4,5)P3 with a maximum at 10 s. This was followed by a rapid decline in Ins(1,4,5)P3 within 30 s to a plateau level above baseline, which gradually declined to reach a new steady state, which was significantly higher than resting Ins(1,4,5)P3 at 30 min. Li+ had no effect on Ins(1,4,5)P3 in resting cells, as well as on the acetylcholine-dependent peak of Ins(1,4,5)P3. However, Li+ caused a transient reduction (at 45 s), followed by a long lasting increase in the Ins(1,4,5)P3 (30 min), as compared with controls. The Li+ effects were dose-dependent and were observed at concentrations used in the treatment of bipolar disorders. Supplementation with inositol had no effect on the level of Ins(1,4,5)P3, at least over the time periods studied. Stimulation of muscarinic receptors with consequent activation of phospholipase C were necessary for the manifestation of Li+ effects in SH-SY5Y cells, Li+ did not interfere with degradation of Ins(1,4,5)P3 after receptor-blockade with atropine, suggesting that Li+ has no direct effect on the Ins(1,4,5)P3-metabolizing enzymes. A direct effect of Li+ on the phospholipase C also is unlikely. Blockade of Ca2+ entry into the cells by Ni2+, or incubation with EGTA, which reduces agonist-stimulated accumulation of Ins(1,4,5)P3, had no effect on the Li(+)-dependent increase in Ins(1,4,5)P3.

Response of peritoneal solid tumours after intraperitoneal chemohyperthermia treatment with cisplatin or carboplatin.

The combination of heat and chemotherapy was studied in an intraperitoneal tumour model. Rats bearing peritoneal CC531 tumours (2-6 mm) were treated i.p. with cDDP or CBDCA [maximal tolerated dose (MTD)] in combination with regional hyperthermia (41.5 degrees C, 1 h) of the peritoneal cavity. The addition of hyperthermia to the i.p. treatment led to a decrease in the MTD of cDDP by 33.3% at 41.5 degrees C. This was due to increased nephrotoxicity. The MTD of CBDCA did not change as a result of hyperthermia treatment. The chemo-hyperthermia treatment resulted in more cDDP or CBDCA DNA adducts in peritoneal tumours after the combined treatment than after chemotherapy alone. The increased tumour platinum concentrations, rising from 1.3 micrograms Pt g-1 tumour at 37 degrees C to 5.4 micrograms Pt g-1 tumour at 41.5 degrees C for cDDP and from 0.2 microgram Pt g-1 tumor to 0.7 microgram Pt g-1 tumour at 41.5 degrees C for CBDCA, contributed considerably to the enhanced numbers of cDDP or CBDCA DNA adducts. As a result of the latter, i.p. chemotherapy combined with regional hyperthermia led to an increase in tumour growth delay (TGD) after increasing the temperature to 41.5 degrees C for cDDP and CBDCA (by 40 days for cDDP, 22 days for CBDCA). These data were in agreement with the in vitro findings, i.e. that higher temperatures led to increased cytotoxicity.

Reduction of cisplatin nephrotoxicity by sodium selenite. Lack of interaction at the pharmacokinetic level of both compounds.

Administration of sodium selenite (Na2SeO3) 1 hr before cis-diamminedichloroplatinum(II) (referred to herein as cisplatin) can protect against the nephrotoxicity of cisplatin. The pharmacokinetic aspects of this interaction were studied in rodents with radiolabeled selenite and cisplatin. Total [75Se]selenium in plasma consisted of [75Se] selenium in plasma proteins and [75Se]selenite in plasma ultrafiltrate. After a short distribution phase, the elimination of [75Se]selenite and total [75Se]selenium proceeded biphasically in the rat, with an initial plasma elimination half-life of [75Se]selenite of 22 +/- 2 min. Coadministration of cisplatin had no effect on the initial nor on the much slower terminal elimination phase of [75Se]selenite nor of total [75Se] selenium. Sodium selenite, in doses protecting against the nephrotoxicity of cisplatin, did not significantly affect areas under the plasma concentration time curve from 0-6 hr nor the initial plasma half-lives of [195mPt]cisplatin (t1/2, 28 +/- 2 min) and total [195mPt]platinum (t1/2, 30 +/- 3 min) in plasma. The much slower terminal elimination phases in plasma and the cumulative urinary excretion of [195mPt] cisplatin and total [195mPt]platinum were neither influenced by sodium selenite. Sodium selenite does not react chemically with cisplatin in vitro. Apparently, bioactivation of selenite is required for its protective effect in vivo. Distribution studies in a mice tumor model indicated that [75Se]selenium is concentrated strongly in the kidney and that the bioactivation of selenite also most likely occurs primarily in the kidneys. We conclude that sodium selenite protects rodents against cisplatin-induced nephrotoxicity without influencing the systemic availability of cisplatin and total platinum.(ABSTRACT TRUNCATED AT 250 WORDS)

Lithium enhances accumulation of [3H]inositol radioactivity and mass of second messenger inositol 1,4,5-trisphosphate in monkey cerebral cortex slices.

We previously reported that lithium, in the presence of acetylcholine, increased accumulations of inositol 1,4,5-trisphosphate and inositol 1,3,4,5-tetrakisphosphate in brain cortex slices from the guinea pig, rabbit, rat, and mouse. In the mouse and rat, the Li(+)-induced increases required supplementation of the medium with inositol. This probably relates to the following facts: (a) Brain cortices of the mouse and rat contain in vivo concentrations of inositol half of that of the guinea pig. (b) Incubated rat brain cortex slices are depleted of inositol by 80%. (c) The slices require 10 mM inositol supplementation to restore in vivo concentrations. We now show that in monkey brain cortex slices, therapeutic concentrations of Li+ increase accumulation of inositol 1,4,5-trisphosphate. The inositol 1,3,4,5-tetrakisphosphate level is not increased. Neither inositol nor an agonist is required. The same effects are seen whether inositol 1,4,5-trisphosphate is quantified by the [3H]inositol prelabeling technique or by mass assay, although mass includes a pool of inositol 1,4,5-trisphosphate that is metabolically inactive. Thus, in a therapeutically relevant model for humans, Li+ increases inositol 1,4,5-trisphosphate levels in brain cortex slices, as was previously seen in lower mammals at non-rate-limiting concentrations of inositol.

A rationale for carboplatin treatment and abdominal hyperthermia in cancers restricted to the peritoneal cavity.

The purpose of this study was to optimize the treatment of cancers restricted to the peritoneal cavity by combining i.p. chemotherapy with abdominal hyperthermia. In vitro experiments demonstrated that the uptake of carboplatin into CC531 tumor cells was increased at temperatures higher than 41.5 degrees C at dose levels of 5 and 50% cell kill. Carboplatin-DNA adduct formation and cytotoxicity, however, were already increased at temperatures of about 40 degrees C, indicating that carboplatin-DNA adduct formation and consequently cytotoxicity could be enhanced by mild hyperthermia (temperatures in the range of 39-41.5 degrees C). CC531 tumor bearing rats were treated i.v. and i.p. with carboplatin (6.15 mg/kg) in combination with regional hyperthermia of the abdomen (41.5 degrees C for 1 h). The mean temperature was 41.5 +/- 0.3 degrees C (SD) in the peritoneal cavity and 40.5 +/- 0.3 degrees C in the esophagus. Enhanced platinum concentrations were found in peritoneal tumors (factor 3) and in kidney, liver, spleen, and lung (a factor 2 average), after the combined i.v. or i.p. carboplatin-hyperthermia treatment. Pharmacokinetic data of i.p. CBDCA combined with hyperthermia demonstrated an increased tumor exposure for total and ultrafiltered platinum in plasma. The areas under the concentration x time curve for total platinum at 37 degrees C and 41.5 degrees C were 69 and 210 microM/h, respectively; for ultrafiltered platinum these values were 47 and 173 microM/h. This may have been due to a slower elimination of platinum from the blood at the higher temperature (t1/2 beta for total platinum 99 and 156 min at 37 and 41.5 degrees C, respectively). The direct exposure of the tumor via the peritoneal fluid appeared to diminish, since the area under the curve for total platinum was lower at 41.5 degrees C than at 37 degrees C (576 microM/h versus 1255 microM/h, respectively). Our results indicate that the advantage of adding hyperthermia is caused by an increased drug exposure of the tumor via the circulation. This was supported by the fact that platinum concentrations in peritoneal tumors after carboplatin treatment at elevated temperatures were similar for the i.p. and i.v. routes.

Li+ increases accumulation of inositol 1,4,5-trisphosphate and inositol 1,3,4,5-tetrakisphosphate in cholinergically stimulated brain cortex slices in guinea pig, mouse and rat. The increases require inositol supplementation in mouse and rat but not in guinea pig.

Li+, beginning at a concentration as low as 1 mM, produced a time- and dose-dependent increase in accumulation of [3H]Ins(1,4,5)P3 and [3H]Ins(1,3,4,5)P4 in acetylcholine (ACh)-stimulated guinea-pig brain cortex slices prelabelled with [3H]inositol and containing 1 mM-inositol in the final incubation period. Similar results were obtained by mass measurement of samples incubated with 10 mM-Li+ by using a receptor-binding assay, although the percentage stimulation of Ins(1,4,5)P3 accumulation by Li+ was somewhat less by this assay. The increase in accumulation of Ins(1,4,5)P3 and Ins(1,3,4,5)P4 by Li+ was absolutely dependent on the presence of ACh. In the absence of added inositol, 1-5 mM-Li+ produced smaller increases in Ins(1,4,5)P3, but the Li(+)-dependent increase in Ins(1,3,4,5)P4 was not as affected by inositol omission. In previous studies with cholinergically stimulated rat and mouse brain cortex slices, Li+ inhibited accumulation of Ins(1,4,5)P3 in rat and inhibited Ins(1,3,4,5)P4 accumulation in rat and mouse [Batty & Nahorski (1987) Biochem. J. 247, 797-800; Whitworth & Kendall (1988) J. Neurochem. 51, 258-265]. We found that Li+ inhibited both Ins(1,4,5)P3 and Ins(1,3,4,5)P4 accumulation in these species, but we could reverse this inhibition by adding 10-30 mM-inositol; we then observed a Li(+)-induced increase in Ins(1,4,5)P3 and Ins(1,3,4,5)P4. The species differences observed in the absence of supplemented inositol were explained by the fact that a much higher concentration of inositol was required to bring the Li(+)-elevated levels of CDP-diacylglycerol (CDPDG) down to baseline in the rat and mouse. These data suggest that inositol is more rate-limiting for phosphatidylinositol synthesis in the presence of Li+ in rat and mouse, which can account for the previous reports of inhibition of Ins(1,4,5)P3 and Ins(1,3,4,5)P4 accumulation by this ion in these species. Thus, in all species examined. Li+ could be shown to increase accumulation of Ins(1,4,5)P3 and Ins(1,3,4,5)P4 in cholinergically stimulated brain cortex slices if the slices were supplemented with sufficient inositol to bring the Li(+)-elevated level of CDPDG down to near baseline, as seen in the absence of Li+. In guinea-pig brain cortex slices, increases in Ins(1,4,5)P3 and Ins(1,3,4,5)P4 accumulation could then be seen at Li+ concentrations as low as 1 mM, which falls within the therapeutic range of plasma concentrations in the treatment of manic-depressive disorders. These observations may have therapeutic implications.

The influence of hyperthermia on the uptake of cisplatin in the rat cervical spinal cord.

The influence of local hyperthermia on the uptake of cisplatin in the rat cervical spinal cord was investigated. After single intraperitoneal or intravenous injection of cisplatin (5 mg/kg body weight), the spinal cord region cervical 5-thoracic 2 was heated for 60 min at mean (S.D.) 41.2 (0.4) degrees C or 40 min 42.4 (0.3) degrees C using a 434 MHz microwave heating device. One day after treatment with either hyperthermia alone, cisplatin alone or the combination, none of the animals expressed neurological symptoms. The spinal cord was dissected and platinum levels were measured by flameless atomic absorption spectroscopy. No difference was found in uptake of platinum in the spinal cord between control- and heat treated animals. In a second series of experiments, the spinal cord was heated for 30-60 min. during a 2 h infusion of cisplatin. One day after treatment at 42.3 degrees C for 60 min, neither motor nor sensory functions were affected and platinum levels did not differ significantly between control and treated animals. Also, platinum levels measured in the spinal cord immediately after cisplatin infusion were not influenced by heat treatment at 42.1 or 43.0 degrees C for 30 min. However, after a heat dose of 60 min 43 degrees C, cisplatin uptake was significantly increased (P less than 0.001) by a factor of 2.8 (1.3). The data demonstrate that mild hyperthermia has no effect on the uptake of cisplatin in the spinal cord, while an injurious heat dose leads to a significant increase in cisplatin uptake. The present findings indicate that, in case of treatment of tumours of the central nervous system with hyperthermia and cisplatin, a treatment which might be toxic for the tumour is well tolerated by the normal nervous tissue.

Optimisation of intraperitoneal cisplatin therapy with regional hyperthermia in rats.

The purpose of this study was to optimise intraperitoneal chemotherapy by combining this modality with regional hyperthermia. In vitro data demonstrated that both the uptake of cisplatin into CC531 tumour cells and cytotoxicity were increased at temperatures of 40 degrees C (factor 4) and 43 degrees C (factor 6) compared to 37 degrees C. The increase of intracellular platinum concentration correlated well with the decrease in survival of these cells. In vivo, rats were treated intraperitoneally with cisplatin (5 mg/kg) in combination with regional hyperthermia of the abdomen (41.5 degrees C, 1 h). The mean (S.D.) temperature in the peritoneal cavity was 41.5 (0.3) degrees C and outside the peritoneal cavity 40.5 (0.3) degrees C. Enhanced platinum concentrations were found in peritoneal tumours (factor 4.1) and kidney, liver, spleen and lung (all around a factor 2.0), after combined cisplatin-hyperthermia treatment. The platinum distribution in peritoneal tumours was more homogeneous after the combined treatment than after cisplatin alone, possibly due to increased penetration of cisplatin into peritoneal tumours. Pharmacokinetic data demonstrated an increased tumour exposure for unfiltered platinum in the peritoneal cavity (area under the curve [AUC] increased from 339 mumol/l/min to 486 mumol/l/min at 37 degrees C and 41.5 degrees C, respectively), and for total and ultrafiltered platinum in the blood. The AUC for total platinum increased from 97.9 to 325.8 mumol/min and for ultrafiltered platinum from 22.2 to 107 mumol/l/min at 37 degrees C and 41.5 degrees C respectively. The latter might be due to a slower elimination of platinum from the blood. The combined treatment, intraperitoneal cisplatin and regional hyperthermia, also increased toxicity. The thermal enhancement ratio (TER) using lethality as endpoint was 1.8.

Selective reduction of cis-diamminedichloroplatinum(II) nephrotoxicity by ebselen.

2-Phenyl-1,2-benzisoselenazol-3(2H)-one (ebselen) is classified as a relatively nontoxic selenium compound, probably because of its bound selenium moiety. In thiol-rich tissues, such as the kidneys, ebselen is converted into selenol intermediates. Selenols are nucleophilic agents which might be able to react with platinum compounds. The influence of ebselen on cis-diamminedichloroplatinum(II) (cisplatin)-induced nephrotoxicity in mice was assessed, using single doses of both compounds. Ebselen prevented cisplatin-induced elevations of blood urea nitrogen and serum creatinine levels and morphological kidney damage in BALB/c mice. This protective effect of ebselen was dose dependent: at a cisplatin dose of 14.5 mg/kg, maximal protection was achieved when a single dose of 10 mg of ebselen/kg was administered 1 h before cisplatin. Administration of ebselen, 10 mg/kg, 1 h after cisplatin also protected against severe nephrotoxicity. Treatment with ebselen did not reduce the antitumor activity of cisplatin against MPC 11 plasmacytoma or Prima breast tumor in BALB/c mice. However, this reduction of cisplatin-induced nephrotoxicity would be of little clinical value if it was achieved at toxic doses of ebselen. Ebselen, 10 mg/kg, did not induce blood urea nitrogen, serum creatinine, serum glutamic pyruvate transaminase, or serum glutamic oxalate elevations in the mice. These results are in agreement with the reported low toxicity of ebselen, which is now in Phase I clinical trials as an antiinflammatory drug. The present results indicate that ebselen may provide protection against cisplatin-induced nephrotoxicity, when it is given before or after cisplatin. This might open new perspectives in cancer chemotherapy.

Selenium-induced protection against cis-diamminedichloroplatinum(II) nephrotoxicity in mice and rats.

The influence of selenium on cis-diamminedichloroplatinum(II) (c-DDP) nephrotoxicity in mice and rats was assessed, using single doses of both compounds. Sodium selenite, 2 mg of selenium per kg, given 1 h before c-DDP, greatly reduced blood urea nitrogen and creatinine levels and morphological kidney damage in both BALB/c mice and Wistar rats, while administration 1 h after c-DDP did not. Liver toxicity of selenium was evaluated by measuring serum glutamic pyruvate transaminase and serum glutamic oxalate transaminase and by routine histology. No liver damage was observed in animals treated with sodium selenite, 2 mg of selenium per kg, and physiological saline or c-DDP. Pretreatment with sodium selenite did not reduce the antitumor activity of c-DDP against MPC 11 plasmacytoma or Prima breast tumor in BALB/c mice. The present results indicate that sodium selenite may provide protection against c-DDP nephrotoxicity, when it is given before c-DDP. Moreover, selenium has an antineoplastic activity against several tumors. The combination of these qualities may open new perspectives in cancer chemotherapy.

Murine mammary tumor response to hyperthermia and radiotherapy evaluated by in vivo 31P-nuclear magnetic resonance spectroscopy.

The response of the s.c.-implanted murine mammary carcinoma NU-82 to hyperthermia was followed as a function of time by 31P-nuclear magnetic resonance spectroscopy. Treatment consisted of elevation of the temperature of the tumors to 41-45 degrees C during 15 min. At 18 h after temperatures of up to 42, 43, 44, and 45 degrees C the ratio of ATP/Pi was unchanged, decreased, largely decreased, and approaching zero, respectively. After the higher doses the relative concentrations (in percentage of total phosphate as visible in the nuclear magnetic resonance spectrum) of phosphomonoesters (mainly phosphoethanolamine) and phosphocreatine also decreased in favor of Pi. The changes in phosphodiesters (mainly glycerophosphocholine) correlated linearly with the changes in ATP (r = 0.84, P less than 0.025). Whereas the limited spectral changes after a dose of 43 degrees C were nullified within 24 h, the more drastic changes after a dose of 45 degrees C lasted at least 8 days. The heavier dose not only induced temporary decreases in tumor perfusion like the lower dose (phase 1) but subsequently, unlike the lower dose, resulted in formation of necrosis (phase 2). In the same tumor we found increases in Pi and decreases in ATP and phosphodiesters after radiotherapy with a dose of 20 Gy. Radiotherapy (20 Gy) combined with hyperthermia (44 degrees C) appeared to strengthen these effects and resulted in an improved tumor response (regression).