Posts Tagged ‘non-linear PK’

Author: Tilda Barliya PhD

In response to the previous post:

Paclitaxel vs Abraxane (albumin-bound paclitaxel)

Pharmacogenomics properties are presented, below.

Paclitaxel is a mitotic inhibitor used in cancer chemotherapy. It was discovered in a U.S. National Cancer Institute program at the Research Triangle Institute (North Carolina)  in 1967 when Monroe E.Wall and Mansukh C.Wani  isolated it from the bark of the Pacific yew tree, Taxus brevifolia and named it taxol. Later it was discovered that endophytic fungi in the bark synthesize paclitaxel.

Paclitaxel is currently being indicated to lung, breast and ovarian cancer as well as  head and neck cancer, and advanced forms of Kaposi’s sarcoma. 

The administration of paclitaxel (Taxol®) through intravenous infusions was achieved by using Cremophor® EL as a vehicle to entrap the drug in micelles and keep it in solution, which affects the disposition of paclitaxel and is responsible for the nonlinear pharmacokinetics of the drug, especially at higher dose levels. (

Although Nonlinear pharmacokinetics (dose-dependented kinetics) may occur in all aspects of pharmacokinetics (absorption, distribution, and/or elimination), it focus on the in the metabolism or MichaelisMenten (MM) kinetics of the drug.

Briefly, it is known that some of these adverse effects such as hypersensitivity reactions were diminished with the administration of corticosteroids and H1 and H2 antihistamine premedication, and by reducing the incidence of grade 3/4 neutropenia with the administration of granulocyte colony-stimulating factors (G-CSF) and shortening paclitaxel infusion time from 24 to 3 h. However, the neurotoxicity, which was believed to be caused by either paclitaxel or Cremophor EL, could not be controlled and became the dose-limiting toxicity of the drug. It was later on found that paclitaxel itself was responsible to the neurotoxicity effects (

Pharmacokinetics and Pharmacodynamics

The selection of pharmacokinetic (PK) parameter end points and basic model types for exposure-toxicity relationships of paclitaxel is usually based on tradition rather than physiological relevance.

pharmacokinetic (PK)-pharmacodynamic (PD) relationships for paclitaxel are still most commonly described with empirically-designed threshold models, which have little or no mechanistic basis and lack usefulness when applied to conditions (eg, schedules, vehicles, or routes of administration) different from those from which they were originally derived. ( As such, the AUC of the unbound paclitaxel is highly important as a pharmacokinetic parameter to describe exposure-neutropenia relationships (the unbound ptx was not evaluated yet). (

The clearance of Cremophor EL in patients was found to be time-dependent, resulting in disproportional increases in systemic exposure being associated with shortening of infusion from 3 hours to 1 hour.

One study (, compare the pharmacokinetics and pharmacodynamics (PD) of paclitaxel between Phase I trials of 3- and 24-h infusions and to determine the most informative pharmacokinetic parameter to describe the PD. The study had 3 main goals

  • (a) to compare the PK and PD of paclitaxel between Phase I studies of 3- and 24-h infusion,
  • (b) to examine the relationship between PK and PD
  • (c) to determine the most informative pharmacokinetic parameter to describe the PD.

Note: Although this study was conducted in ~1993-1995, is has been cited extensively and paved the was to other clinical trials with similar results.

27 patients were treated in a Phase I study of paclitaxel by a 3-h infusion at one of six doses: 105, 135, 180, 210, 240, and 270 mg/m2. Pharmacokinetic data were obtained from all patients. Paclitaxel concentrations were measured in the plasma and urine using HPLC. Similar eligibility criteria were designed for the 24-hr infusion with these doses were 49.5, 75, 105, 135, and 180 mg/m2 . Plasma and urine samples for pharmacokinetic evaluation of paclitaxel were collected.

Pharmacokinetic Analysis: Pharmacokinetic parameters, Cmax, AUC, t112, and MRT were obtained by a noncompartmental moment method. Cmax was actually observed peak concentration. AUC and MRT were computed by trapezoidal integration with extrapolation to infinite time.

Pharmacodynamic Analysis: The pharmacokinetic/pharmacodynamic relationships were modeled with the sigmoid maximum effect


Pharmacokinetic analysis:

The drug plasma concentration increased throughout the 3-h infusion period and began to decrease immediately upon cessation of the infusion with t112 of 9.9-16.0 h and MRT of 6.47-10.24 h (Fig. 1). Both Cmax and AUC increased with increasing doses (r = 0.865, P <0.001 for Cmax r 0.870, P < 0.001 for AUC), although the pharmacokinetic behavior appeared to be nonlinear (Fig. 2). The mean Cmax and AUC at a dose of 270 mg/m2 were more than 3-fold greater than those at a dose of 135 mg/m2. CL and V, decreased with increasing doses (Table 1). The urinary excretion of paclitaxel over 75 h was less than 15% of the dose administered, which indicated that non-renal excretion is the primary route of drug elimination.

The urinary excretion of paclitaxel over 75 h was less than 15% of the dose administered, which indicated that non-renal excretion is the primary route of drug elimination.

Comparison of PD between 3-h and 24-h Infusion

Groups. AUC and duration of plasma concentration (h) above (7>) 0.05-0.1 LM correlated with the % D in granulocytes with p values less than 0.05. The best parameter predicting granulocytopenia was T> 0.09 pM with the minimum of the Akaike Information Criterion. In the 24-h schedule, dose, AUC, and T > 0.04-0.07 pM were demonstrated to correlate with the % D in granulocytes. The best parameter predicting granulocytopenia in the 24-h schedule was T > 0.05 p.M.

Nonhematological toxicities such as peripheral neuropathy, hypotension, and arthralgialmyalgia mainly observed in the 3-h infusion group had no relationship with Cm or AUC which are much higher in the 3-h infusion group, although peripheral neuropathy and musculoskeletal toxicity have been suggested to be associated with AUC on a 6- (12) or 24-h (29) schedule.


In the past, the major adverse effects encountered with Taxol were severe hypersensitivity reactions, mainly attributed to Cremophor EL; hematologic toxicity, primarily appearing in the form of severe neutropenia; and neurotoxicity, mainly seen as cumulative sensory peripheral neuropathy. The mechanism for the neurotoxicity has been demonstrated to involve ganglioneuropathy and axonopathy caused by dysfunctional microtubules in dorsal root ganglia, axons and Schwann cells.

Variability in paclitaxel pharmacokinetics has  been associated with the adverse effects of the  drug. Thus, polymorphisms in genes encoding  paclitaxel-metabolizing enzymes, transporters and therapeutic targets have been suggested  to contribute to the interindividual variability in toxicity and response.

Further characterization of  genes involved in paclitaxel elimination and drug  response was performed, including the identification of their most relevant genetic variants. The organic anion transporting polypeptide (OATP)  1B3 was identified as a key protein for paclitaxel hepatic uptake and polymorphisms in the genes encoding for paclitaxel metabolizing enzymes and transporters (CYP2C8, CYP3A4) CYP3A5, P-glycoprotein and OATP1B3) (

***It is important to note that  the allele frequencies for many of these polymorphisms are subject to important ethnicity  specific differences, with some alleles exclusively present in specific populations (e.g., the Caucasian CYP2C8*3).

For the CYP2C8 gene, two alleles common in Caucasians that result in amino acid changes CYP2C8*3 (R139K; K399R) and CYP2C8*4 (I264M), were described. The former has been shown to possess an altered activity, while the latter does not seem to have functional
consequences. In addition, two CYP2C8 haplotypes were recently shown to confer an increased and reduced metabolizing activity, respectively.

CYP3A5 was found to be highly polymorphic owing to CYP3A5*3, CYP3A5*6 and CYP3A5*7 , with the latter two being African-specific polymorphisms.

Pharmacogenetic studies comparing the most relevant polymorphisms in these genes and paclitaxel pharmacokinetics have rendered contradictory results, with some studies finding no associations while others reported an effect for ABCB1, CYP3A4 or CYP2C8 polymorphisms on specific pharmacokinetic parameters.

Again, with respect to paclitaxel neurotoxicity risk, some studies have rendered positive results for ABCB1 , CYP2C8  and CYP3A5  polymorphisms, while others found no significant associations.

Note: These differences might be caused by underpowered studies and by differences in the patients under study.

Changes affecting microtubule  structure and/or composition have been shown to affect paclitaxel efficacy, probably by reducing drug–target affinity. Mainly, resistance to tubulin-binding agents has been associated with an overexpression of b-tubulin isotype III,
which seems to be caused by a deregulation of the microRNA family 200.

However, the clinical utility of these findings remains to be established; furthermore, the identification of biomarkers that could be used to individualize paclitaxel treatment remains a challenge.

In summary,

  1. Pharmacokinetics: Paclitaxel seems to have a non-linear (=dose-dependent) PK parameters.
  2. Pharmcokinetics- Pharmacodynamics: Previous clinical trials did NOT take into account the unbound concentrations of Ptx and therefore in the PK analysis, therefore newly designed clinical trials should take that into consideration. This is very important since the neurotoxicity is attributed to ptx and not its vehicle Cremophor (as shown in the PD analysis)
  3. Difficult to compare between the 3hr and 24hr infusion schedule as most clinical trials did NOT used similar dose-regime making the comparison very hard.
  4. Pharmacogenetics: Different polymorphisms seems to attribute to the been suggested  to contribute to the interindividual variability in toxicity and response.
  5. Prospective pharmacogenetic-guided clinical trials will be required in order to accurately establish the utility of the identified markers/strategies for patients and healthcare systems.



Read Full Post »

Author: Tilda Barliya PhD

Taxanes, are  diterpenes produced by the plants of the genus Taxus (yews), and are widely used as chemotherapy agents. Taxane agents include paclitaxel (Taxol) and docetaxel (Taxotere). The taxane class of drugs inhibit the microtubules by stabilizing GDP-bound tubulin in the microtubule, thereby inhibiting the process of cell division. Paclitaxel (trade name Taxol) is dissolved in Cremophor EL and ethanol, as a delivery agent and much of the clinical toxicity of paclitaxel is associated with the solvent Cremophor EL in which it is dissolved.

Albumin-bound paclitaxel (trade name Abraxane, also called nab-paclitaxel) is an alternative formulation where paclitaxel is bound to albumin nano-particles (particle size of approximately 130 nanometers). nab-Paclitaxel utilises the natural properties of albumin to reversibly bind paclitaxel, transport it across the endothelial cell and concentrate it in areas of tumour. The proposed mechanism of drug delivery involves, in part, glycoprotein 60-mediated endothelial cell transcytosis of paclitaxel-bound albumin and accumulation in the area of tumor by albumin binding to SPARC (secreted protein, acidic and rich in cysteine).

When evaluating paclitaxel vs the albumin-bound paclitaxel in Pharmacokinetics (PK) clinical trials, few important questions are raised:

  • What is the total paclitaxel?
  • How much  FREE  paclitaxel is generated by each type of drug (Taxol vs Abraxane)?
  • Do they have a linear or non-linear PK curves?


Few differences between Taxol (paclitaxel) and Abraxane (albumin-bound paclitaxel) are:

  • Time of administration; Taxol (3hrs) and Abraxane (30min)
  • PK curves; Taxol (non-linear and therefore less predictable) and Abraxane (linear and therefore more predictable)
  • Doses; Taxol (175 mg/m2) and Abraxane (260 mg/m2)

These differences affect the analysis of the results obtained from many clinical trials conducted in multiple clinical centers and need to be taken into consideration.

In 2006: single arm phase II safety study was conducted to support the approval of adjuvant breast cancer. The FDA published the Clinical PK Comparison of Total Paclitaxel Study c008-0
Sparreboom A. et al  Clin Cancer Res 2005; 11:4136-4143
Study Design:
  • Randomized, Phase 3, open label
  • Sample size: 460 patients
  • 70 sites: Russia (77%), UK (15%), Canada and US (9%)
  • 2 Arm: Abraxane 260 mg/m2 as a 30-minute infusion and Taxol 175 mg/m2  as a 3-hour infusion
  • 59% second line or greater and 77% previous anthracycline exposure
  • Designed to show non inferiority in RR

(mean ± %CV)


260 mg/m2



175 mg/m2















22969 3543 6.5 x 89 20 4.4 X


14789 12603 1.17 x 57 72 0.80 x


21 15 1.43 x


21 15 1.43 x




664 433 1.53 x


664 433 1.53 x


FREE paclitaxel was NOT measured!!!!

Toxicity profile:

  • Taxol has a higher incidence of neutropenia and hypersensitivity reactions
  • Abraxane has a higher incidence of peripheral neuropathy, nausea, vomiting, diarrhea and asthenia

Overall Survival:

  • There was no difference in overall survival between the Abraxane and Taxol treatment groups. HR (Abraxane/Taxol) was 0.90, p=0.348 (log rank).
  • No conclusions can be drawn from a subgroup analysis when the main analysis was not statistically significant.
  • Multiple subgroup analyses using different criteria without p value adjustments
  • P-values are not interpretable


In the presentation at the American Society of Clinical Oncology (ASCO) meeting in Chicago, many eyebrows have been raised over Abraxane vs Paclitaxel study ( 05/Eyebrows_raised_at_ASCO_over_Abraxane_vs_paclitaxel_study.aspx)

The Phase III study enrolled 799 patients with locally advanced or metastatic breast cancer who were randomised to receive one of the three therapies – paclitaxel (the standard of care), Abraxane (nanoparticle albumin bound -‘nab’ – paclitaxel) or Ixempra (ixabepilone) – on a weekly basis with each cycle consisting of three weeks of treatment followed by a one-week break. Some 98% of patients also received Roche’s Avastin (bevacizumab), which had its approval for breast cancer revoked by the US Food and Drug Administration in November 2011.

The data from the study, presented at ASCO by lead investigator Hope Rugo at the University of California, San Francisco, stated that median progression-free survival was 10.6 months for those receiving paclitaxel, 9.2 months for nab-paclitaxel, and 7.6 months for ixabepilone.       Abraxane was NO better than paclitaxel !  The major surprise was over the  150mg high does chosen for the Abraxane arm, well above the 100mg for which Abraxane is approved in over 40 or so countries,

However, when searching the literature and evaluating multiple publications, Abraxane seems to be more efficacious over Taxol

Benefits of Abraxane vs. Taxol or Onxal are:
– more effective at treating tumors because a higher dosage can be delivered.
– decrease in side effects from solvent related hypersensitivity reactions.
– decreased use of medications to combat the solvent related hypersensitivity reactions.
– decreased time of administration.


In summary,

Abraxane (the albumin-bound paclitaxel) seems to have better benefits over the free paclitaxel as stated above. However, due to the differences in PK properties and lack of FREE drug measurements, more clinical studies needs to be conducted in order the understand the true values and differences between the two drug.




Read Full Post »