, Volume 48, Issue 5, pp 794–847


A Review of its Pharmacodynamic and Pharmacokinetic Properties and Therapeutic Potential in the Treatment of Cancer


  • Caroline M. Spencer
    • Adis International Limited
  • Diana Faulds
    • Adis International Limited
Drug Evaluation

DOI: 10.2165/00003495-199448050-00009

Cite this article as:
Spencer, C.M. & Faulds, D. Drugs (1994) 48: 794. doi:10.2165/00003495-199448050-00009



Paclitaxel is a new anticancer agent with a novel mechanism of action. It promotes polymerisation of tubulin dimers to form microtubules and stabilises microtubules by preventing depolymerisation.

In noncomparative trials, continuous infusion of paclitaxel 110 to 300 mg/m2 over 3 to 96 hours every 3 to 4 weeks produced a complete or partial response in 16 to 48% of patients with ovarian cancer and 25 to 61.5% of patients with metastatic breast cancer, many of whom were refractory to treatment with cisplatin or doxorubicin, respectively. 23 to 100% of patients with ovarian cancer achieved complete or partial responses with paclitaxel in combination with cisplatin, carboplatin, cyclophosphamide, altretamine and/or doxorubicin. Similarly, response rates of 30 to 100% were observed with paclitaxel plus doxorubicin, cisplatin, mitoxantrone and/or cyclophosphamide in patients with metastatic breast cancer. Comparative trials in patients with advanced ovarian cancer showed paclitaxel therapy to produce greater response rates than treatment with parenteral hydroxyurea (71 vs 0%) or cyclophosphamide (when both agents were combined with cisplatin) [79 vs 63%]. Paclitaxel was also more effective than mitomycin in 50 patients with previously untreated breast cancer (partial response in 20 vs 4% of patients).

Paclitaxel therapy also produced promising results in patients with advanced squamous cell carcinoma of the head and neck, malignant melanoma, advanced non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), germ cell cancer, urothelial cancer, oesophageal cancer, non-Hodgkin’s lymphoma or multiple myeloma, and was successfully combined with cisplatin, carboplatin and/or etoposide in patients with NSCLC, SCLC or advanced squamous cell carcinoma of the head and neck.

Hypersensitivity reactions were initially a concern with administration of paclitaxel, although current dosage regimens have reduced the incidence of these events to less than 5%. The major dose-limiting adverse effects of paclitaxel are leucopenia (neutropenia) and peripheral neuropathy. Other haematological toxicity was generally mild. Cardiac toxicity was reported in small numbers of patients and most patients developed total alopecia.

Several aspects of paclitaxel use remain to be clarified, including the optimal treatment schedule and infusion time, confirmation of the tolerability profile and efficacy of combination regimens in an expanded range of malignancies. Long term follow-up of paclitaxel recipients will also allow the effects of the drug on patient survival to be determined. Nevertheless, paclitaxel is a promising addition to the current therapies available, with significant activity reported in patients with advanced ovarian or breast cancer or other types of tumour. The drug has activity as salvage or initial therapy, alone or in combination with currently used agents, against a number of cancers. In particular, paclitaxel should be considered a first-line agent in women with ovarian cancer refractory to cisplatin and as a second-line agent for the treatment of metastatic breast cancer if doxorubicin is ineffective.

Pharmacodynamic Properties

Paclitaxel, a new anticancer agent first isolated from the bark of Taxus brevifolia, has a wide spectrum of antineoplastic activity. It has in vitro cytotoxicity against human ovary, breast, cervical, pancreas, prostate, head and neck, colon, gastric, bladder, lung and CNS cancers, melanoma, hepatoma and leukaemia cell lines, often at concentrations lower than those achieved in the serum of patients. Similarly, fresh human lung, ovarian, breast and endometrial cancer cells, sarcoma and leukaemia cells are sensitive to paclitaxel. Increasing exposure time to paclitaxel appears to increase the activity of the drug. However, resistant tumour cell lines and fresh tumour isolates have been reported.

In general, activity of paclitaxel in vitro was greater than that of tiazofurine, cisplatin, etoposide, doxorubicin or fluorouracil against human tumours and similar or less than (generally 2 to 4 times lower) that of docetaxel, as measured by mean concentrations producing 50% inhibition of cell growth. In vitro, synergistic interactions generally occurred when exposure to paclitaxel was initiated up to 48 hours prior to cisplatin, when cells were exposed to doxorubicin prior to incubation with higher concentrations of paclitaxel, when cells pretreated with edatrexate were exposed to paclitaxel, when 24-hour exposure to paclitaxel preceded incubation with melphalan, thiotepa or fluorouracil or when paclitaxel was combined with tiazofurine, calcitriol, vinorelbine or estramustine. However, the order in which drugs were added appears to be an important determinant of synergism or antagonism. Paclitaxel also enhances the effects of irradiation against many cell lines in vitro and a mammary carcinoma in mice. It has also been evaluated in combination with irradiation in patients with non-small cell lung cancer (NSCLC) or brain tumours.

Three or 4 weeks’ treatment with intraperitoneal or subcutaneous paclitaxel 12.5 mg/kg for 4 or 5 consecutive days out of seven induced total tumour regression in nude mice with human glioblastoma (9 to 20%), breast cancer (80%), lung cancer (14%) or tongue cancer xenografts (17%). No regressions were observed in mice with a fast-growing gliosarcoma or endometrial or ovarian cancer xenografts. In other studies, intravenously or intraperitoneally administered paclitaxel was active against intraperitoneal human ovarian and prostate cancer xenografts transplanted into mice, but was inactive in 3 of 4 subcutaneous human pancreatic adenocarcinoma models.

The mechanisms of inherent or acquired resistance to paclitaxel have not yet been fully elucidated, but increased expression of the multidrug resistant gene (mdr-1) and alterations in α- or β-tubulin have been implicated. Clinically relevant concentrations of polyoxyethylated castor oil (polyoxyl 35 castor oil), the base used clinically in formulations of paclitaxel, reverses resistance of some cells to the drug, as does co-incubation with verapamil, quinidine or cyclosporin. In mice, definite cross-resistance developed in amsacrine-resistant leukaemia cell lines, marginal cross-resistance developed in doxorubicin-, dactinomycin- or mitoxantrone-resistant cell lines, but no cross-resistance was noted in camptothecin-, melphalan-, cisplatin-, cytarabine- or methotrexate-resistant cell lines. Similarly, no cross-resistance to cisplatin was noted in ovarian or lung cancer cell lines, although cross-resistance to doxorubicin and etoposide did occur in the lung cancer cell lines.

Reversible neurotoxicity develops in some patients treated with paclitaxel. Neurological examination shows the most common abnormality to be reduced or absent ankle reflexes. When paclitaxel is injected directly into the sciatic nerve of rats, microtubules proliferate in both axons and Schwann cells at the expense of most other organelles and result in congestion and dilation of Schwann cell bodies and development of significant stretches of naked axons, some of which display unusual axon-Schwann cell relationships. Remyelination appears dependent on decreasing axon diameter. Apparent normalisation of Schwann cells, axons and endoneural cells generally occurs 6 months after injection of paclitaxel in animals.

Paclitaxel also produces concentration-dependent suppression of human peripheral blood mononuclear and natural killer cell cytotoxicity against cancer cell lines. These effects are reduced by pretreatment with interleukin-2. In contrast, stimulated, but not unstimulated, release of the proinflammatory cytokines interleukin-1β and tumour necrosis factor-α from human mononuclear phagocytes is dose-dependently enhanced by paclitaxel. Microtubule-associated functions of neutrophils are also inhibited by the drug.

Pharmacokinetic Properties

Plasma concentrations of paclitaxel increase throughout 6- or 24-hour infusions and begin to decline immediately upon cessation of the infusion. Although maximum plasma concentration (Cmax) and area under the concentration-time curve (AUC) were dose-related in small numbers of patients who received paclitaxel 120 to 300 mg/m2 as a 3-/6- or 24-hour continuous infusion, the pharmacokinetic behaviour of paclitaxel appears to be nonlinear. Administration of paclitaxel 135 to 350 mg/m2 produces mean steady-state plasma drug concentrations higher [0.20 to 8.54 mg/L (0.23 to 10 μmol/L)] than concentrations producing antimicrotubule effects in vitro (at least 0.1 μmol/L). Paclitaxel is cleared rapidly from plasma initially, eliminated over a prolonged period (4.3 to 49.76 hours), and is extensively protein bound (88 to 98%). The apparent volume of distribution is large and is correlated with administered drug dose. Paclitaxel does not appear to easily cross the blood-brain barrier in humans. In rodents, paclitaxel is predominantly distributed to the liver, lung, spleen, adrenal and salivary glands, heart, muscle, kidneys, stomach, intestine and pancreas but not the nervous system or testes.

Peak intraperitoneal concentrations of 16 to 277 mg/L (19 to 324 μmol/L) occur 30 to 60 minutes after completion of administration of intraperitoneal paclitaxel 25 to 175 mg/m2. Clearance from the peritoneal cavity is slow (mean clearance of 0.42 L/m2/day and half-life of 73.4 hours), and peritoneal exposure to paclitaxel appears to be 336 to 2890 times greater than systemic exposure.

Mean total body clearance ranges from 8.04 to 23.55 L/h/m2 following a 3- to 24-hour infusion of paclitaxel 15 to 275 mg/m2 and is not correlated with dose. Hepatic metabolism and biliary clearance appear to be the major means of elimination of paclitaxel. Five metabolites of paclitaxel have been identified in human bile, of which 2 are monohydroxylated and 1 is a dihydroxylated derivative. The major metabolite of paclitaxel isolated in human liver microsomes is 6μ-hydroxytaxol. Urinary clearance of paclitaxel was minimal (16% or less). Clearance of paclitaxel was reduced in 9 patients with liver tumours and aspartate aminotransferase levels at least 1.5 times normal, compared with values in 13 patients without hepatic involvement (20.16 vs 28.26 L/h/m2). Dialysis did not appear to significantly alter pharmacokinetic parameters of paclitaxel in 1 patient. The pharmacokinetic profile of paclitaxel in children does not appear to differ from that observed in adults.

Mean clearance rates of paclitaxel are reduced by prior cisplatin administration compared with the reverse schedule (19.26 vs 24.3 L/h/m2). Concomitant administration of doxorubicin and paclitaxel did not alter steady-state concentration or clearance of either drug compared with monotherapy. Fluconazole and ketoconazole inhibited, and cimetidine or erythromycin hade little or no effect, on the metabolism of paclitaxel in vitro.

The severity and incidence of adverse effects appeared to be correlated with AUC, steady-state concentration, the duration plasma paclitaxel concentrations remained above a certain level and absolute dose of paclitaxel in some, but not all, studies.

Therapeutic Potential

In phase II studies, paclitaxel was usually administered as a 24-hour continuous infusion to patients with a good or excellent performance status. However, 3-hour infusions have been successfully used in patients with advanced ovarian or breast cancer. Noncomparative trials showed continuous intravenous infusion of paclitaxel 110 to 300 mg/m2 over 3, 24 or 96 hours every 3 to 4 weeks to produce a complete or partial response in 16 to 48% of patients with ovarian cancer and 20 to 61.5% of patients with metastatic breast cancer. The median time to disease progression was longer in patients with ovarian or breast cancer treated with paclitaxel 175 versus 135 mg/m2 and in those with ovarian cancer treated with a 3- versus 24-hour infusion. Median survival was 9 months in 652 evaluable patients with ovarian cancer treated with 24-hour infusions of paclitaxel 135 mg/m2 every 3 weeks. In general, patients with ovarian cancer responded to treatment with paclitaxel 135 to 300 mg/m2 after a median of 9.2 weeks and duration of response ranged from 2 to 18.6 (median 7.2) months. Intraperitoneal paclitaxel produced a limited response in a small number of patients with major manifestations of residual ovarian or breast cancer confined to the peritoneal cavity enrolled in a phase I trial.

Many patients with ovarian or breast cancer who responded to therapy with paclitaxel had been refractory to cisplatin or doxorubicin therapy, respectively. Response was not correlated with estrogen receptor status in patients with breast cancer.

Phase I studies conducted in small numbers of patients showed paclitaxel in combination with cisplatin and cyclophosphamide to produce objective responses in 58% of patients with newly diagnosed advanced ovarian cancer and with cisplatin to produce complete or partial responses in all 5 patients with suboptimally debulked ovarian cancer. Objective responses were also achieved by patients with previously treated ovarian cancer who received paclitaxel and cyclophosphamide (47%), paclitaxel and oral altretamine (23%) or paclitaxel and doxorubicin 25 mg/m2 (40%); however, only small numbers of patients were enrolled in these studies. Paclitaxel produced greater clinical response rates than treatment with parenteral hydroxyurea (71 vs 0%), and when combined with cisplatin, it was more effective than cyclophosphamide plus cisplatin (79 vs 63%) in patients with ovarian cancer. In the latter study, median duration of progression-free survival was 17.9 months for paclitaxel recipients and 13.8 months for women treated with cyclophosphamide.

In small numbers of patients with breast cancer, combination therapy with paclitaxel and doxorubicin produced response rates of 44.5 to 100%. Paclitaxel plus cyclophosphamide produced a partial response in 30 or 62% of patients with metastatic disease and the combination of paclitaxel and cisplatin produced partial responses in 52 or 94% of patients.

Paclitaxel, alone or in combination with cisplatin, produced complete or partial responses in 21 or 37% and 33% of patients, respectively, with advanced or metastatic squamous cell carcinoma of the head and neck. Treatment with paclitaxel also resulted in objective responses in patients with oesophageal cancer (32%), non-Hodgkin’s lymphoma (42%) or previously untreated multiple myeloma (23%). Promising results were obtained with 24-hour infusions of paclitaxel 200 to 275 mg/m2 in patients with malignant melanoma (response rates of 12 to 33%), cisplatin-resistant germ cell cancer (response rate of 24%), previously untreated advanced transitional cell carcinoma of the urothelium (response rate of 42%), advanced previously untreated NSCLC [response in 21 or 24% patients, median survival of 24.1 or 40 weeks and 1-year survival rate of 42%] or previously untreated small cell lung cancer (partial responses in 34 or 40.5%). However, paclitaxel was less effective in patients with previously treated NSCLC. When combined with cisplatin or carboplatin, paclitaxel produced complete or partial responses in 25 to 50% and 22 to 80%, respectively, of small numbers of patients with NSCLC. Combination therapy with paclitaxel, cisplatin and etoposide was very promising in 7 patients with previously untreated NSCLC; 6 patients had a partial response after 2 courses of therapy. Similarly, all 8 patients with SCLC achieved a complete or partial response after treatment with paclitaxel, carboplatin and etoposide.


Severe acute hypersensitivity reactions occur in some patients treated with paclitaxel, in many cases necessitating therapeutic intervention and/or discontinuation of therapy. However, the incidence of these reactions has been reduced to less than 5% subsequent to the introduction of premedication with dexamethasone and histamine H1- and H2-receptor antagonists. Hypersensitivity reactions frequently include dyspnoea, hypotension, manifestations of angioedema, urticaria, flushing and/or erythematous rash and usually occur during administration of the first or second dose of paclitaxel, in most instances within 10 minutes of initiating the drug infusion.

The dose-limiting toxicity of paclitaxel is generally neutropenia or leucopenia, although administration of granulocyte colony-stimulating factor (G-CSF) and a reduction in infusion time from 24 to 3 hours reduces the incidence of this toxicity. In clinical trials, grade 4 leucopenia or granulocytopenia developed in 16 to 100% of patients treated with paclitaxel 135 to 250 mg/m2 over 24 hours. As the duration of neutropenia is usually only 3 to 10 days and is not cumulative, myelosuppression is not dose limiting in many patients. However, 21 to 62% of patients who received the drug as a 24-hour infusion were hospitalised because of febrile neutropenia, while sepsis during neutropenia resulted in death in about 1.3% of patients. Neutropenia tended to worsen when cisplatin therapy preceded a 24-hour infusion of paclitaxel, or when a 24-hour infusion of paclitaxel was administered prior to doxorubicin or cyclophosphamide, when compared with the reverse schedules. Although grade 3 or 4 thrombocytopenia and anaemia occurs infrequently with paclitaxel 135 to 350 mg/m2 (0 to 24% of patients), and in some studies thrombocytopenia did not occur, many patients develop mild manifestations of this toxicity.

Dose-related peripheral neuropathy can also be dose-limiting, with grade 3 or 4 toxicity affecting 2 to 27% of patients treated with paclitaxel. Neuropathy follows a ‘stocking and glove’ distribution, mainly involves neurosensory manifestations and tends to be symmetrical. However, in isolated instances patients have developed motor abnormalities. Peripheral neurotoxicity tended to occur more frequently in patients with pre-existing neuropathy or other risk factors for neuropathy in a small number of trials. Equivocal results were obtained with respect to the cumulative nature of paclitaxel-induced neuropathy. Myalgia or arthralgia also develop 2 to 6 days after administration of paclitaxel in 14 to 100% of patients. Although these symptoms resolve over 2 to 7 days, they can be extremely distressing and dose limiting in some patients. Optic nerve damage has also been reported in 9 of 47 patients receiving a 3-hour infusion of paclitaxel 175 or 225 mg/m2 in 1 trial.

Paclitaxel causes transient asymptomatic bradycardia in some patients, but serious dysrhythmias rarely develop. However, significant cardiovascular events occurred in small numbers of patients treated with paclitaxel in clinical trials. These included myocardial infarction, atrial fibrillation, mild congestive heart failure, ventricular and supraventricular tachycardia, ventricular arrhythmia, Wenckebach syndrome, sudden death, hypotension, asymptomatic T-wave inversion, atrioventricular block and asymptomatic sinus arrest.

Other adverse effects associated with paclitaxel include total alopecia (in almost all patients), grade 3 or 4 mucositis (1.4 to 30% of patients), predominantly mild nausea and vomiting and local venous toxicity (4 to 64%). Many additional usually mild effects (taste impairment, diarrhoea, anorexia, sore throat, stomatitis, fatigue, fever, oedema, hypotension, hypomagnesaemia, dyspnoea, headache, facial flushing, phlebitis, transient azotaemia and transient hepatocellular dysfunction) have also been reported in patients receiving the drug. In addition, grand mal seizures, unexplained reversible renal insufficiency, typhlitis (when paclitaxel was combined with doxorubicin) or cutaneous radiation recall reaction have been reported rarely. Age, prior therapy or the total cumulative dose of paclitaxel received appeared to have little effect on the tolerability profile of the drug.

Dosage and Administration

As paclitaxel is poorly soluble in water, it is formulated in a vehicle of 50% polyoxyethylated castor oil and 50% alcohol (ethanol).

Although adequate dose-response trials have not been completed for paclitaxel, a dosage of 135 to 175 mg/m2, administered as a 3- or 24-hour intravenous infusion every 3 weeks, appears to be effective in patients with metastatic ovarian cancer. A 3-hour infusion of paclitaxel 175 mg/m2 appears effective in patients with metastatic breast cancer after failure of previous antineoplastic chemotherapy. Subsequent courses of paclitaxel should be administered only when neutrophil and platelet counts are adequate, and dosage should be reduced by 20% if patients experience severe neutropenia or severe peripheral neuropathy.

In clinical trials, higher doses of paclitaxel could be administered with G-CSF without an increase in the incidence of haematological toxicity, but the incidence of peripheral neuropathy increased. If a 24-hour infusion of paclitaxel is to be administered with cisplatin, paclitaxel should be administered first. Although recommendations are not available concerning the use of paclitaxel in elderly patients or children, clinical studies have included patients aged from 2 to 88 years, many of whom were elderly, and used no special dosages in these patients. However, children tolerated higher dosages of paclitaxel than adults in dose-ranging studies.

Recommendations state that all patients should receive premedication with oral dexamethasone (or clemastine), and intravenous diphenhydramine and cimetidine or ranitidine before paclitaxel to reduce the risk of hypersensitivity reactions. Continuous cardiac monitoring is not necessary during treatment with paclitaxel, except in patients with serious pre-existing conduction abnormalities. Although few data are available concerning the use of paclitaxel in patients with liver impairment, lower dosages of the drug should be administered to these patients.

Copyright information

© Adis International Limited 1994