Cerebrospinal fluid concentrations of carboplatin in a patient without blood-brain barrier disruption

Pharmacokinetic considerations in the treatment of CNS tumours

Despite aggressive therapy, the majority of primary and metastatic brain tumour patients have a poor prognosis with brief survival periods. This is because of the different pharmacokinetic parameters of systemically administered chemotherapeutic agents between the brain and the rest of the body. Specifically, before systemically administered drugs can distribute into the CNS, they must cross two membrane barriers, the blood-brain barrier (BBB) and blood-cerebrospinal fluid (CSF) barrier (BCB). To some extent, these structures function to exclude xenobiotics, such as anticancer drugs, from the brain. An understanding of these unique barriers is essential to predict when and how systemically administered drugs will be transported to the brain. Specifically, factors such as physiological variables (e.g. blood flow), physicochemical properties of the drug (e.g. molecular weight), as well as influx and efflux transporter expression at the BBB and BCB (e.g. adenosine triphosphate-binding cassette transporters) determine what compounds reach the CNS. A large body of preclinical and clinical research exists regarding brain penetration of anticancer agents. In most cases, a surrogate endpoint (i.e. CSF to plasma area under the concentration-time curve [AUC] ratio) is used to describe how effectively agents can be transported into the CNS. Some agents, such as the topoisomerase I inhibitor, topotecan, have high CSF to plasma AUC ratios, making them valid therapeutic options for primary and metastatic brain tumours. In contrast, other agents like the oral tyrosine kinase inhibitor, imatinib, have a low CSF to plasma AUC ratio. Knowledge of these data can have important clinical implications. For example, it is now known that chronic myelogenous leukaemia patients treated with imatinib might need additional CNS prophylaxis. Since most anticancer agents have limited brain penetration, new pharmacological approaches are needed to enhance delivery into the brain. BBB disruption, regional administration of chemotherapy and transporter modulation are all currently being evaluated in an effort to improve therapeutic outcomes. Additionally, since many chemotherapeutic agents are metabolised by the cytochrome P450 3A enzyme system, minimising drug interactions by avoiding concomitant drug therapies that are also metabolised through this system may potentially enhance outcomes. Specifically, the use of non-enzyme-inducing antiepileptic drugs and curtailing nonessential corticosteroid use may have an impact.

Plasma and cerebrospinal fluid pharmacokinetics of intravenous oxaliplatin, cisplatin, and carboplatin in nonhuman primates

PURPOSE

Describe and compare the central nervous system pharmacology of the platinum analogues, cisplatin, carboplatin, and oxaliplatin and develop a pharmacokinetic model to distinguish the disposition of active drug from inert platinum species.

EXPERIMENTAL DESIGN

Oxaliplatin (7 or 5 mg/kg), cisplatin (2 mg/kg), or carboplatin (10 mg/kg) was given i.v. Serial plasma and cerebrospinal fluid (CSF) samples were collected over 24 hours. Plasma ultrafiltrates were prepared immediately. Platinum concentrations were measured using atomic absorption spectrometry. Areas under the concentration x time curve were derived using the linear trapezoidal method. CSF penetration was defined as the CSF AUC(0-24)/plasma ultrafiltrate AUC(0-24) ratio. A four-compartment model with first-order rate constants was fit to the data to distinguish active drug from inactive metabolites.

RESULTS

The mean +/- SD AUCs in plasma ultrafiltrate for oxaliplatin, cisplatin, and carboplatin were 61 +/- 22, 18 +/- 6, and 211 +/- 64 micromol/L hour, respectively. The AUCs in CSF were 1.2 +/- 0.4 micromol/L hour for oxaliplatin, 0.56 +/- 0.08 micromol/L hour for cisplatin, and 8 +/- 2.2 mumol/L hour for carboplatin, and CSF penetration was 2.0%, 3.6%, and 3.8%, respectively. For oxaliplatin, cisplatin, and carboplatin, the pharmacokinetic model estimated that active drug accounted for 29%, 79%, and 81% of platinum in plasma ultrafiltrate, respectively, and 25%, 89%, and 56% of platinum in CSF, respectively. The CSF penetration of active drug was 1.6% for oxaliplatin, 3.7% for cisplatin, and 2.6% for carboplatin.

CONCLUSIONS

The CSF penetration of the platinum analogues is limited. The pharmacokinetic model distinguished between active drug and their inactive (inert) metabolites in plasma and CSF.

Paclitaxel and carboplatin in the treatment of small-cell lung cancer patients resistant to cyclophosphamide, doxorubicin, and etoposide: a non-cross-resistant schedule

PURPOSE

To evaluate the efficacy of paclitaxel and carboplatin (PC) in small-cell lung cancer (SCLC) patients resistant to cyclophosphamide, doxorubicin, and etoposide (CDE).

PATIENTS AND METHODS

We performed a phase II study with PC in SCLC patients who relapsed within 3 months after first-line treatment with CDE. Paclitaxel administration (175 mg/m2 by a 3-hour intravenous infusion) was followed by a 30-minute infusion of carboplatin (area under the curve 7; Chatelut formula) once every 3 weeks for five cycles. Dexamethasone, clemastine, and ranitidine were standard premedication before every cycle.

RESULTS

Included were 35 patients (median age, 59 years; 16 with limited disease and 19 with extensive disease; Eastern Cooperative Oncology Group performance status of < or = 1; median time off treatment 6 weeks) who were previously treated with CDE (n = 33), oral etoposide (n = 2), and reinduction CDE (n = 15); only one patient had received three CDE treatments of five cycles. The CDE regimen was followed by local thoracic radiotherapy in seven patients. Hematologic toxicity of grade 3 or 4, for leukopenia was 27% and 6%, for thrombocytopenia 21% and 13%, and for anemia 17% and 0%, respectively, for a total of 132 cycles. Two patients had neutropenic fever; no toxic death occurred. Nonhematologic toxicity was paresthesia CTC grade 3, diarrhea grade 4, and myalgia grade 3 in one patient each. Reversible paresthesia (CTC grade 1 and 2) in toes and fingers was reported in 69% of patients. Thirty-four patients were assessable for response: complete response in two patients, partial response in 23 patients, stable disease in eight patients, and progressive disease in one patient (response rate, 73.5%; 95% confidence interval, 59% to 88%). One patient was found to have atypical carcinoid at pathologic review and was excluded. Median time to progression was 21 weeks (range, 3 to 40 weeks). Median survival was 31 weeks (range, 6 to 112 weeks). One-year survival was 9%.

CONCLUSION

Second-line PC in CDE-resistant SCLC patients yields a high response rate and seems non-cross-resistant to CDE. Toxicity was mild in these poor-prognosis patients.