The first 2 authors contributed equally to this work.
Farnesyl transferase expression determines clinical response to the docetaxel-lonafarnib combination in patients with advanced malignancies†
Article first published online: 1 MAR 2011
Copyright © 2011 American Cancer Society
Volume 117, Issue 17, pages 4049–4059, 1 September 2011
How to Cite
Kauh, J., Chanel-Vos, C., Escuin, D., Fanucchi, M. P., Harvey, R. D., Saba, N., Shin, D. M., Gal, A., Pan, L., Kutner, M., Ramalingam, S. S., Bender, L., Marcus, A., Giannakakou, P. and Khuri, F. R. (2011), Farnesyl transferase expression determines clinical response to the docetaxel-lonafarnib combination in patients with advanced malignancies. Cancer, 117: 4049–4059. doi: 10.1002/cncr.26004
We thank Dr. Anthea Hammond for assistance in editing, Dr. Wade M. Smith for protocol writing, Dr. Mourad for assistance in data analysis, and Dr. Selwyn Hurwitz for critical reading of the manuscript.
- Issue published online: 19 AUG 2011
- Article first published online: 1 MAR 2011
- Manuscript Accepted: 10 JAN 2011
- Manuscript Revised: 10 DEC 2010
- Manuscript Received: 11 OCT 2010
Lonafarnib (LNF) is a protein farnesyl transferase (FTase) inhibitor that has shown synergistic activity with taxanes in preclinical models and early stage clinical trials. Preclinical findings suggested tubulin acetylation and FTase expression levels may be important determinants of drug sensitivity that would help identify patient populations more likely to benefit from this regimen. This pilot study evaluated the biological effects of LNF and docetaxel (DTX) combination therapy in refractory solid tumors by comparing pretreatment and post-treatment tumor biopsies.
Patients with histologically confirmed locally advanced or metastatic solid malignancies refractory to standard therapies or with no effective therapies available were eligible. Patients were randomized to 1 of 4 dosing cohorts: 1) 30 mg/m2, 100 mg; 2) 36 mg/m2, 100 mg; 3) 30 mg/m2, 150 mg; or 4) 36 mg/m2, 150mg of DTX intravenously weekly, LNF orally twice daily, respectively.
Of the 38 patients enrolled, 36 were treated, and 29 were evaluable for toxicity and response assessment. The combination of LNF and DTX was tolerated in all cohorts with the exception of a 28% incidence of grade 3/4 diarrhea, which was manageable with aggressive antidiarrheal regimens. Seven patients derived clinically meaningful benefit from this combination treatment; these patients had significantly lower basal FTase-beta mRNA expression levels than the mean study population level (P < .05). Correlation of clinical benefit with tubulin acetylation content as well as basal acetyl-tubulin content were evaluated. However, no significant correlation was found.
Despite the small number of patients, these findings support our preclinical mechanistic studies and warrant further clinical investigations using FTase-beta mRNA expression as a potential predictive biomarker to select for an enriched patient population to study the effects of taxane and FTase inhibitor combination therapies. Cancer 2011. © 2011 American Cancer Society.
Lonafarnib (SCH 66,336) is a small molecule inhibitor of farnesyl transferase (FTase), which adds a 15-carbon farnesyl group to several G-proteins important in intracellular signaling involved in cell survival, including Ras, RhoB, Pxf, Rap2, and cyclic GMP phosphodiesterase.1-3 Mutations in the Ras family of oncogenes are common in human cancers4 and have been associated with shortened survival in several human tumor types.5, 6 Because Ras farnesylation was found to be required for its membrane localization and thus its oncogene activation,7-9 lonafarnib and other FTase inhibitors (FTIs) were developed as potential Ras inhibitors and were shown to inhibit Ras function.3, 10, 11 However, farnesylation of other proteins has also been shown to be involved in the antitumor effects of lonafarnib and other FTIs.12-22 Despite its modest single-agent activity, lonafarnib has shown highly synergistic activity in combination with taxanes in preclinical models and early stage clinical trials.23-34
Members of our team reported the initial phase 1 trial of lonafarnib in combination with paclitaxel and determined the recommended phase 2 doses to be lonafarnib 100 mg twice a day and paclitaxel 175 mg/m2 every 21 days.35 The subsequent phase 2 trial of lonafarnib and paclitaxel in taxane-refractory nonsmall cell lung carcinoma (NSCLC) reported promising antitumor activity, a partial response (PR) rate of 10% (3 of 29 patients) and a stable disease (SD) rate of 38% (11 of 29 patients).30 The ability to overcome taxane resistance would have significant clinical impact given the wide range of neoplastic diseases treated with taxane-based therapy. However, the phase 2 trial did not provide data that could be used to elucidate the biological basis for overcoming taxane resistance.
Our group also demonstrated that lonafarnib not only synergizes with microtubule-stabilizing taxanes in vitro, but it also is able to reverse taxane resistance in drug-resistant cancer models.24, 25 Mechanistically, we have shown lonafarnib synergizes with taxanes through inhibition of tubulin deacetylase, histone deacetylase 6 (HDAC6), which leads to increased tubulin acetylation, microtubule stability, and enhanced taxane binding to microtubules.24, 25 In addition, we have shown that FTase physically associates with HDAC6 and microtubules (docetaxel's target) and that FTase knockdown sensitizes cells to lonafarnib/taxane drug combination.36 Taken together, these preclinical findings suggested tubulin acetylation and basal FTase expression levels may be important determinants of drug sensitivity that would help us identify patients more likely to benefit from the combination of lonafarnib and docetaxel.
The primary goal of this study was to evaluate the biological effects of combination therapy with lonafarnib and docetaxel in refractory solid tumors by comparing pretreatment and post-treatment tumor biopsies. Tumor specimens were evaluated for effective drug-target engagement, biologic interactions, surrogate markers of biologic activity, and potential predictive markers of benefit. Secondary goals were to determine safety, tolerability, toxicity profile, and preliminary evidence of antitumor activity of the combination of lonafarnib and docetaxel.
Patients were randomized into 4 cohorts of 9 patients with different doses of lonafarnib and docetaxel. Within each cohort, patients were randomly subdivided into 3 separate subgroups receiving their post-treatment biopsy after docetaxel alone, lonafarnib alone, and docetaxel/lonafarnib combined. This randomization scheme would potentially allow us to identify what impact, if any, different doses of the FTI/taxane combination have on microtubule stabilization and farnesyl transferase inhibition.
MATERIALS AND METHODS
Eligibility criteria included histologically confirmed locally advanced or metastatic solid malignancies refractory to standard therapies; age ≥18 years; tumor accessible for repeat biopsy; ECOG performance status of ≤2; life expectancy ≥12 weeks; discontinuation of potent CYP3A4 inducers/inhibitors; adequate laboratory values including leukocyte count ≥3000 cells/mm3, absolute neutrophil count ≥1500 cells/mm3, platelet count ≥100,000/mm3, hemoglobin ≥9.0 g/dL, total bilirubin level less than or equals to the upper limit of normal (ULN), albumin ≥2.5 g/dL, aspartate aminotransferase or alanine aminotransferase ≤2 × ULN, prothrombin time, and partial thromboplastin time ≤1.5 × ULN. Women of childbearing potential were required to have a negative pregnancy test. Exclusion criteria included greater than grade 2 neuropathy and inability to swallow pills. All study subjects signed an informed consent form approved by the Emory University Institutional Review Board.
Patients were enrolled from April 2006 to April 2008. The primary objective was to determine the molecular interaction between docetaxel and lonafarnib in tumor samples. Secondary objectives included 1) determination of the safety and toxicity of docetaxel in combination with lonafarnib, 2) determination of pharmacokinetic interactions, 3) determination of molecular interactions between docetaxel and lonafarnib in peripheral blood mononuclear cells (PBMC).
Study subjects were randomly assigned (Fig. 1) to 1 of 4 dosing cohorts, 1) 30 mg/m2, 100 mg; 2) 36 mg/m2, 100 mg; 3) 30 mg/m2, 150mg; or 4) 36 mg/m2, 150 mg of docetaxel intravenously weekly, lonafarnib orally twice daily, respectively. Lonafarnib capsules were administered twice daily, and docetaxel was administered weekly for 3 weeks every 28 days. Premedication for docetaxel included a 5-HT3 antagonist and dexamethasone.
All patients underwent pretreatment tumor biopsies 1 week before drug administration. By using a second randomization, patients were randomly assigned to 1 of 3 schedules for a second tumor biopsy. Group 1 was treated with docetaxel alone on day 1, repeat biopsy on day 2, and then initiation of lonafarnib treatment. Group 2 was treated with lonafarnib alone on day 1, repeat biopsy before dosing on day 5, and then docetaxel initiation. Group 3 was treated with lonafarnib on day 1, docetaxel on day 4, and repeat biopsy on day 5, before lonafarnib.
Dose modification was permitted. A single 25% dose reduction of docetaxel due to toxicity was permitted, and all subsequent treatments were administered at the reduced dose. Grade 3 or 4 neutropenia (without fever) with recovery before the next planned cycle did not require dose modification. Doses of docetaxel were also held for abnormal liver function, alkaline phosphatase >2 × ULN, bilirubin >ULN.
A single dose reduction of lonafarnib due to toxicity was permitted. All subsequent treatments were at the reduced dose (150 mg twice daily was reduced to 100 mg twice daily; 100 mg twice daily was reduced to 100 mg in the morning and 50 mg in the evening). Doses of lonafarnib were held for grade 3 or 4 thrombocytopenia until platelet counts returned to ≥100,000/mm3. Grade 3 or 4 nausea, vomiting, or diarrhea required discontinuation of lonafarnib until toxicities returned to grade 1 or better, and subsequent doses were reduced by 1 level.
Study subjects were removed from the study for the following reasons: 1) patient request, 2) progressive disease, 3) unacceptable toxicity, 4) investigator judgment.
Baseline tumor evaluation with cross-sectional imaging was performed within 4 weeks of therapy initiation, and subsequent imaging was performed every 8 weeks. Tumor response was determined by using the Response Evaluation Criteria In Solid Tumors (RECIST) criteria.
Blood samples (6 mL) were collected in sodium heparinized at times 0 (pre), 1, 1.25, 1.75, 4, 7.5, and 24 hours after docetaxel infusion. Lonafarnib maximum steady-state concentrations, Css(max), were collected for each patient based on twice daily administration for 3 or 4 days before docetaxel administration. Samples were immediately centrifuged at 3000 rpm for 15 minutes at 4°C; plasma was removed, separated into 3 aliquots, and stored at −70°C.
Lonafarnib concentrations were analyzed at Taylor Technology (Princeton, New Jersey) using liquid chromatographic method with tandem mass spectrometric detection (LC-MS/MS) with a lower limit of quantification (LLQ) of 5 ng/mL and a linear concentration range of 5-2500 ng/mL. Docetaxel concentrations were assayed at Pharmaceutical Product Development (PPD, Richmond, Virginia) by LC-MS/MS with a LLQ of 10 ng/mL (10-5000 ng/mL).
Individual pharmacokinetic parameters for docetaxel were calculated from plasma concentration-time curves by WinNonlin v 5.2 (Pharsight, Mountain View, California) using noncompartmental methods. Parameters reported included area under the concentration-time curve extrapolated to infinity (AUCinf), terminal half-life (t1/2), clearance (CL), and volume of distribution (Vd). Lonafarnib Css(max) values are reported because of inconsistent acquisition of other time points in the sampling scheme.
Each tumor biopsy specimen was divided into 3 portions and processed as follows: flash frozen, embedded in paraffin, and the remainder saved for microtubule analysis. Samples for microtubule analysis were immediately fixed with complete PHEMO buffer.37
Tissue Microtubule Integrity and Stability Assessment by Immunofluorescence Staining Followed by Confocal Microscopy
For immunofluorescence processing, tumor biopsies were processed, imaged, and analyzed.37 Images were acquired using a Zeiss 5LIVE confocal microscope with a 63×/1.3 NA objective (Carl Zeiss MicroImaging GmgH, Germany).
Immunofluorescence of Acetylated Tubulin in PBMCs
PBMCs were centrifuged at 500 RPM for 5 minutes onto poly-lysine–treated cover slips then fixed in PHEMO buffer (68 mM PIPES, 25 mM HEPES, pH 6.9, 15 mM EGTA, 3 mM MgCl2, 10% [vol/vol] DMSO).23-25, 38, 39 Confocal z-sections were acquired using a Zeiss LSM510 META microscope23, 38, 39 and analyzed using Metamorph 6.0 (Carl Zeiss MicroImaging GmgH, Germany).
Immunohistochemistry for Acetylated Tubulin
Immunohistochemistry for acetylated tubulin was performed at the Molecular Cytology Core Facility of Memorial Sloan-Kettering Cancer Center using a Discovery XT processor (Ventana Medical Systems, Tucson, Arizona) and a monoclonal antiacetylated tubulin antibody (Sigma clone 6-11B-1; Sigma-Aldrich, St. Louis, Missouri), followed by biotinylated mouse secondary antibody (Vector Labs, Burlingame, California), streptavidin-HRP D (Ventana Medical Systems) and DAB Detection Kit (Ventana Medical Systems). Stained tissue sections were mounted on glass slides and scanned using the ZEISS MIRAX SCAN. TIFF images of tissue sections were obtained using MIRAX Viewer Software. The tumor area in each slide was delineated by a pathologist and levels of tubulin acetylation were assessed in the defined tumor area using Metamorph. The threshold level of acetylated tubulin was determined, and the intensity of the resultant image was expressed as a percentage of the total intensity in the tumor area .
FTase Alpha and Beta mRNA Expression
Total RNA was isolated from fresh frozen tumor biopsies using TRIzol reagent (Invitrogen, Carlsbad, California). cDNA was prepared from 1 μg total RNA by using random primers and High-Capacity cDNA Reverse Transcription Kit from Applied Biosystems (Foster City, California). Real-time PCR reactions were prepared using SYBR Green Supermix (BioRad, Hercules, California). Primers for FTase alpha were 5′-TGATCGTGCTGTATTGGAGAG-3′ and 5′-CTGTGCTGTGTTTGCTTTGAA-3′; primers for FTase beta were 5′-TACTATTGCCCTCCATCTTCCTCC-3′ and 5′-GACCTCTTGGATCTTTTCTTCTAC-3′; primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were 5′-GGAGTCAACGGATTTGGTCG-3′ and 5′-CTTGATTTTGGAGGGATCTCG-3′. Quantification was performed using Applied Biosystems (ABI) Prism 7700 Real-time qPCR system under the following cycling conditions: (step 1) 50˚C (2 minutes); (step 2) 95˚C (10 minutes); (step 3) 40 cycles of 95˚C (15 seconds), and 60˚C (1 minutes). The fluorescence threshold value was determined using SDS and RQ Manager software. The relative expression level of FTase alpha or FTase beta was normalized using GAPDH as an internal standard (Applied Biosystems, Carlsbad, Calif). The Ct value for each FTalpha and FTbeta expression was determined and normalized to the Ct value obtained with GAPDH from the same samples. The mean of all normalized FTase alpha or beta expression values from all samples was set to 1.
FTase expression was dichotomized into low or high expression for each patient based on the median of sample delta Ct value. Kaplan-Meier product limit estimators were generated and plotted for progression-free survival (PFS) stratified by FTase expression levels (low vs high). A log-rank test was performed. PFS was defined as the time interval from the start of treatment to disease progression.
For each patient, the change in acetylated tubulin staining was categorized as having either increased or decreased on the basis of the change of acetylated tubulin levels before and after treatment. Kaplan-Meier product limit estimators stratified by change in acetylated tubulin were generated for PFS. The log-rank test was performed.
Of 38 patients enrolled, 36 were treated, and 29 were evaluable for toxicity and response assessment (Table 1). All patients underwent tumor biopsies; however, because of technical reasons (eg, insufficient tumor sample in biopsy specimen), several patients did not have adequate paired biopsies for molecular analysis.
|Head & neck||8|
|Small bowel adenocarcinoma||1|
|Anus squamous cell carcinoma||1|
With the exception of diarrhea, which was manageable with aggressive antidiarrheal medications, the combination of lonafarnib and docetaxel was tolerated in all 4 cohorts (Table 2). The most common drug-related clinical adverse events were diarrhea (all grades, 69%; grade 3/4, 28%), nausea (all grades, 61%; grade 3/4, 14%), vomiting (all grades, 56%; grade 3/4, 8%), fatigue (all grades, 47%; grade 3/4, 22%), neuropathy (all grades, 38%; grade 3/4, 3%), weight loss (all grades, 37%; grade 3/4, 0%), and rash (all grades, 17%; grade 3/4, 3%).
|Toxicity||Incidence per CTC Grade|
|3 No.||4 No.||5 No.|
The most common drug-related laboratory abnormalities were hyperglycemia (all grades, 92%; grade 3/4, 23%), hyponatremia (all grades, 53%; grade 3/4, 11%), hypoalbuminemia (all grades, 53%; grade 3/4, 16%), leukopenia (all grades, 53%; grade 3/4, 19%), anemia (all grades, 50%; grade 3/4, 14%), hypocalcemia (all grades, 47%; grade 3/4, 12%), hypokalemia (all grades, 39%; grade 3/4, 6%), creatinine elevation (all grades, 34%; grade 3/4, 3%), elevated amylase (all grades, 30%; grade 3/4, 6%), hypophosphatemia (all grades, 25%; grade 3/4, 0%), and thrombocytopenia (all grades, 25%; grade 3/4, 9%). Hyperglycemia was a common finding and most likely due to protocol mandated dexamethasone premedication before docetaxel infusion. All episodes of hyperglycemia were treated with either oral agents (ie, glyburide, metformin) or subcutaneous insulin. Drug-induced leukopenia resolved within 1 week of delayed treatment.
Three deaths occurred during the study. One patient died because of aspiration pneumonia in the setting of chemotherapy-induced neutropenia. A second patient died because of complications of urosepsis in the setting of chemotherapy-induced neutropenia. The third patient death was due to rapidly progressing community-acquired pneumococcal sepsis, which was not thought to be related to study treatment because the patient's neutrophil counts were in the normal range.
Clinical benefit was defined as SD, PR, or complete response (CR). Seven patients clinically benefitted from treatment with docetaxel and lonafarnib (Table 3). One patient with parotid carcinoma had a CR, and 6 patients had SD lasting from 6 to 10 months. Of note, several of the patients who benefitted from protocol therapy had documented disease progression when previously treated with taxanes, 1 patient with prior docetaxel treatment and 5 patients with prior paclitaxel treatment.
|Patient No.||Disease||Best Response||Months of Benefit||Prior Taxane (Dose)||Best Response to Prior Taxane|
|5||Adenoid cystic carcinoma of parotid||SD||10||Paclitaxel (175 mg/m2 q21d)||SD|
|12||MFH||SD||8||Docetaxel (75 mg/m2 q21d)||PD|
|20||SCC lung||SD||4||Paclitaxel (175 mg/m2 q21d)||SD|
|22||Adenoid cystic carcinoma of nasopharynx||SD||8||Paclitaxel (175 mg/m2 q21d)||PD|
|28||Parotid carcinoma||CR||10||Paclitaxel (60 mg/m2 qwk)||SD|
|30||Bronchoalveolar||SD||6||Paclitaxel (175 mg/m2 q21d)||PD|
For docetaxel, the AUC0-inf and t1/2values were only reported for patients in whom concentrations were above the LLQ at 24 hours, allowing for a full set of values for parameter calculations (n = 13; Fig. 2A). Comparing docetaxel exposure in the presence and absence of lonafarnib at steady-state concentrations, a trend toward increased docetaxel exposure as measured by Cmax and AUC0-inf values in patients receiving the combination was seen; however, this did not correspond with greater clinical efficacy or toxicity. Median docetaxel dose-normalized AUC0-inf was compared between groups receiving and not receiving lonafarnib (Fig. 2B) and showed no difference in mean AUC0-inf values, despite a numeric trend (P = .46). Lonafarnib Cssmax values were variable, but mean values increased in a dose-dependent fashion.
Tubulin acetylation analysis in pretreatment and post-treatment biopsies
Tubulin acetylation content was determined as the percentage of cells staining positive for tubulin acetylation. Because of the need to validate multiple biomarkers and limited tissue supply, we were able to identify only evaluable tumor areas with matched pretreatment and post-treatment biopsies in 9 patients. In 2 patients who clinically benefitted, 1 showed a robust increase in tubulin acetylation (60% at baseline to 100% post-treatment) while the second did not show a significant change (70% to 80%) (Figs. 3 and 4). Among the 7 patients who did not clinically benefit, 3 patients showed no increase in the levels of tubulin acetylation, while 4 patients had a 20%-70% increase. We also investigated whether basal acetyl-tubulin content could serve as a predictive biomarker of benefit. However, all evaluable pretreatment samples from patients who benefitted had levels of tubulin acetylation of greater than 40%, as did those patients who did not benefit.
FTase expression in pretreatment biopsies
Quantitative real-time polymerase-chain reaction (RT-PCR) for FTase-alpha and FTase-beta mRNA was performed in 28 baseline tumor samples. Results were normalized to GAPDH expression and categorized into low or high expression for each patient based on the median of sample delta CT values. Kaplan-Meier analysis indicated that patients with low FTase-alpha expression had a trend toward improved survival, albeit not significant (P = .1145) (Fig. 5A). Log-rank test showed that the difference in PFS between patients with low versus high FTase-beta expression was statistically significantly different (P < .05; Fig. 5B).
Tubulin acetylation analysis in PBMCs
Pretreatment and post-treatment levels of acetylated tubulin in PBMCs did not correlate with PFS (P = .4986).
This National Cancer Institute (NCI P01)-funded, translational, biomarker-driven, clinical trial investigated the interaction between docetaxel and lonafarnib, evaluating molecular predictors of outcome as well as pharmacokinetic and pharmacodynamic interactions. Our previous clinical and mechanistic studies suggested lonafarnib could reverse taxane resistance.24, 30 However, the biological basis of this finding was not well understood. Mechanistically, we have shown that the synergy between taxanes and FTIs could be explained by FTIs ability to inhibit the tubulin deacetylase function of HDAC6, leading to microtubule stabilization, enhanced tubulin acetylation, and taxane binding.24, 25, 28, 36
Of the 36 patients treated, 7 patients derived benefit from protocol treatment. Remarkably, 6 of these patients had previously failed taxane-based therapy; thus, the lonafarnib/docetaxel combination was able to overcome this resistance. The pharmacokinetic analysis of both agents was consistent with previous reports.40-42 Although dose-normalized docetaxel exposure in patients receiving lonafarnib was numerically higher, this difference was not significant (P = .46). In the 7 patients who benefitted, no difference in exposure was seen compared with those who did not benefit, suggesting factors other than plasma drug concentrations contributed to the likelihood of benefit.
Tubulin acetylation was assessed as a biomarker indicative of effective drug-target engagement and predictive of a patient's response to this combination therapy. Inconsistencies were noted between the increases observed in tubulin acetylation in post-treatment biopsies and the actual clinical response of individual patients, which were attributable to the small number of biopsies with analyzable tumors areas and the high baseline level of tubulin acetylation. Because most subjects had failed prior taxane chemotherapy, it is possible that tubulin acetylation analysis was biased by prior treatment. When examining tumor biopsy specimens as well as PBMCs for changes in levels of acetylated tubulin, immunofluorescence did not reveal a statistically significant correlation with PFS. Interestingly, microtubule disruption, evinced by microtubule bundling or aberrant mitosis, did not always lead to cell death nor was it associated with clinical response (data not shown). This suggests additional pathways downstream of microtubule targeting are dysregulated, potentially through the overexpression of antiapoptotic proteins like Bcl2 or deficient spindle assembly checkpoints.43
Recent studies from our group have revealed cells with a stable knockdown of FTase were sensitized to taxane and lonafarnib alone or in combination.36 Similar results were obtained with FTase-beta knockdown, leading us to hypothesize that patients with lower FTase expression at baseline could potentially show a better response to this combination. Indeed, patients that benefitted from treatment had statistically significantly lower basal mRNA expression levels of FTase-beta compared with the mean mRNA expression levels for the study population. Lower basal mRNA expression levels of FTase-alpha were also seen in the 7 patients who clinically benefited; however, this association did not reach statistical significance. These clinical observations are in agreement with our previous studies in which knockdown of FTase-alpha resulted in concomitant downregulation of FTase-beta subunit and vice versa because the 2 subunits are cotranslationally regulated. However, in the clinical samples, we did not observe such a coordinated expression for the 2 subunits, suggesting that in patients, additional pathways may affect the expression of each subunit individually. Our data from the current trial also show FTase-beta may be a more accurate biomarker than FTase-alpha for predicting clinical benefit to lonafarnib/docetaxel, possibly due to the finding that FTase and geranylgeranyl transferase exist as alpha and beta heterodimers and share a common alpha subunit but a homologous beta subunit (25% sequence identity). Therefore, FTase-beta would be expected to be a more specific biomarker for FTase activity than FTase-alpha.44, 45 Moreover, mutations in FTase that confer resistance to lonafarnib have been described at residue betaY361 both in vitro and in patients46; the presence of such a mutation could explain the lack of clinical benefit from the lonafarnib/docetaxel combination.
During the conception of this trial, we set the ambitious goal of prospectively collecting tumor samples from all participants before and after treatment. Although we had a well thought out plan, unforeseen technical issues precluded us from obtaining adequate amounts of viable tumor tissue for all our planned correlative studies. Many of our samples contained necrotic debris or little malignant tissue limiting our ability to complete all planned testing. Given the small number of paired biopsies, we were unable to investigate the impact of administering lonafarnib versus docetaxel versus the combination, had on molecular endpoints. Nor were we able to definitively correlate pretreatment and post-treatment acetylation levels with clinical benefit. In future studies investigating molecular endpoints, we suggest having a pathologist at bedside during biopsy to confirm the presence of adequate tumor cells so as to improve tumor collection rates and quality of specimens collected.
Overall, the regimen of docetaxel and lonafarnib appears to be tolerable. The most common toxicities of diarrhea, nausea, and vomiting were mostly grade 1/2 and manageable with oral regimens. We do acknowledge a moderate (28%) incidence of grade 3/4 diarrhea, but with aggressive antidiarrheal medications, we found the trial regimen tolerable. Hyperglycemia was most likely related to dexamethasone premedication but was manageable with oral hypoglycemics or insulin. Of the 3 deaths that occurred during the study, 2 were study-related (sepsis in the setting of chemotherapy-induced neutropenia); the third death was thought to be unrelated to study treatment (rapidly progressing community-acquired pneumococcal sepsis). There is no evidence that lonafarnib enhanced docetaxel-induced neutropenia; it bears mentioning that the study population was extensively pretreated. During the course of the trial, enrollment was suspended after each patient death, and the clinical data were reviewed by the Emory University, Winship Cancer Institute Data Safety Monitoring Board (WCI-DSMB). Accrual was reopened only after the WCI-DSMB found the risks to participants were acceptable.
This study highlights the feasibility and potential utility of incorporating serial tumor biopsies into clinical trials. As in previous trials, investigating the combination of lonafarnib and taxanes, we observed several patients with clinical benefit despite pretreatment with taxanes in prior regimens. However, unlike previous studies, we were able to identify mRNA expression levels of FTase-beta and FTase-alpha as potential predictive biomarkers. This study also stresses the importance of National Cancer Institute funding through the P01 mechanism to support correlative translational research to go from bedside to bench (early clinical trials of taxanes and lonafarnib) and then back to bedside (mechanistic studies on lonafarnib and taxane synergy) to confirm preclinical observation of mechanisms of FTI-enhanced effects of taxanes.3, 24, 25, 30, 35, 36, 47, 48 Although the agent, lonafarnib, is unlikely to be developed further by the pharmaceutical industry, the data gathered during the conduct of this study may prove to be a valuable step forward in the personalization of cancer care.
CONFLICT OF INTEREST DISCLOSURES
This study was supported by NIH P01 CA116676 to Fadlo R. Khuri; funds from the Georgia Cancer Coalition (GCC) to scholars Fadlo R. Khuri, Suresh S. Ramalingam, Dong M. Shin, and Adam Marcus; a grant from the American College of Clinical Pharmacy (ACCP) and Pharmaceutical Product Development (PPD) to R. Donald Harvey; and a grant from Sanofi-Aventis to Michael P. Fanucchi and Fadlo R. Khuri. R. Donald Harvey received the PPD Bioanalytical Fluid and Tissue Sample Grant Award from the American College of Clinical Pharmacy (ACCP) Research Institute and PPD, for analysis of docetaxel concentrations.