Preclinical Investigation of Nanoparticle Albumin-Bound Paclitaxel as a Potential Treatment for Adrenocortical Cancer
Michael J. Demeure, MD, MBA,*+ Elizabeth Stephan, PhD, ; Shripad Sinari, MS,; David Mount, PhD,} Steven Gately, PhD,; Paul Gonzales, PhD,t Galen Hostetter, MD,; Richard Komorowski, MD, § Jeff Kiefer, PhD,t Clive S. Grant, MD, | Haiyong Han, PhD, ; Daniel D. Von Hoff, MD,*} and Kimberly J. Bussey, PhDt
Background: Traditional drug discovery methods have a limited role in rare cancers. We hypothesized that molecular technology including gene expres- sion profiling could expose novel targets for therapy in adrenocortical carci- noma (ACC), a rare and lethal cancer. SPARC (secreted protein acidic rich in cysteine) is an albumin-binding matrix-associated protein that is proposed to act as a mechanism for the increased efficacy of a nanoparticle albumin-bound preparation of the antimicrotubular drug Paclitaxel (nab-paclitaxel).
Methods: The transcriptomes of 19 ACC tumors and 4 normal adrenal glands were profiled on Affymetrix U133 Plus2 expression microarrays to identify genes representing potential therapeutic targets. Immunohistochemical anal- ysis for target proteins was performed on 10 ACC, 6 benign adenomas, and 1 normal adrenal gland. Agents known to inhibit selected targets were tested in comparison with mitotane in the 2 ACC cell lines (H295R and SW-13) in vitro and in mouse xenografts.
Results: SPARC expression is increased in ACC samples by 1.56 ± 0.44 (µ ± SD) fold. Paclitaxel and nab-paclitaxel show in vitro inhibition of H295R and SW-13 cells at IC50 concentrations of 0.33 pM and 0.0078 p.M for paclitaxel and 0.35 p&M and 0.0087 µM for nab-paclitaxel compared with mitotane concentrations of 15.9 uM and 46.4 p.M. In vivo nab-paclitaxel treatment shows a greater decrease in tumor weight in both xenograft models than mitotane.
Conclusions: Biological insights garnered through expression profiling of ACC tumors suggest further investigation into the use of nab-paclitaxel for the treatment of ACC.
(Ann Surg 2012;255:140-146)
A drenocortical carcinoma (ACC) is one of the deadliest of human cancers. Although rare, with a worldwide incidence of 1 to 2 per million, the prognosis is poor with a 5-year survival rate of 10% to 20%, because of the typically late presentation and the limited effectiveness of broad-spectrum chemotherapy.1 At present, the only realistic opportunity for cure is complete surgical removal, but un- fortunately, metastatic spread is already present in 40% to 70% of patients at the time of diagnosis.2 First approved in 1960, mitotane (Bristol-Myers Squibb Company, New York) remains the standard chemotherapeutic treatment for this disease.3 Mitotane, also known as o,p’-DDD, is a derivative of the pesticide DDT and an adrenolytic.
Response rates are poor with mitotane as a single agent (positive re- sults in approximately 22%), but survival for those who do respond is improved from 14 to 50 months.4 For most patients, mitotane is poorly tolerated because of its severe toxic side effects, including obliteration of the healthy contralateral adrenal gland.5 Etoposide, doxorubicin, and cisplatin in combination with mitotane, known as the Italian regi- men, have been reported to produce clinical response rates up to 50% in advanced ACC patients, although complete remissions are rare.6 Anthracycline agents and other inhibitors of topoisomerase II, in- cluding doxorubicin, have been used in the treatment of patients with ACC, but the efficacy of these agents is limited because of expression of the multidrug resistance P-glycoprotein, encoded by the ABCB1 gene in adrenocortical cancers.7,8 Recent advances in genomic profil- ing now offer new opportunities to expose vulnerabilities that could allow for the development of novel treatments that are sorely needed for ACC.
Pharmacological targeting of the microtubules in patients with ACC has shown only limited success. The western yew extract pacli- taxel exerts it activity by promoting tubulin polymerization and stabi- lization of microtubules resulting in arrest at G2-M in the cell cycle and induction of apoptosis.9 Paclitaxel has shown promise in the labo- ratory setting but has not been studied in patients with ACC. Paclitaxel was reported to inhibit in vitro growth of H295R, an adrenocortical cancer cell line, but there have been no subsequent investigations re- garding the use of this agent for ACC.1º Abraxane®, or nab-paclitaxel, is a novel albumin stabilized, cremophor-free, nanoparticle formula- tion of paclitaxel.11 Purported benefits of this formulation included less toxicity due to the absence of cremophor, a caster oil derivative that is used to make paclitaxel soluble in water. Cremophor results in hypersensitivity reactions in patients who must be pretreated with corticosteroids. The drug nab-paclitaxel is bound to nanoparticles of albumin, a protein that the body uses to transport water-insoluble compounds. It has been postulated that increased levels of drug may get into tumor cells through transendocytosis via active transport by gp60 on the endothelial cell surface into the interior of the tumor.12 In a phase III trial involving 454 patients with metastatic breast cancer, the overall response rate to nab-paclitaxel was 33% compared with 19% for the standard preparation of paclitaxel. The median response rate was also longer at 21.9 weeks compared with 16.1 weeks.13
The extracellular matrix protein SPARC (secreted protein acidic rich in cysteine) modulates cell adhesion and growth and is widely implicated in angiogenesis and tumor progression.14,15 Over- expression of SPARC has been associated with many cancers in- cluding those of the breast, pancreas, esophagus, stomach, colon, prostate, kidney, and in melanoma.16,17 SPARC binds to albumin, including the albumin of nab-paclitaxel, and consequently the pacli- taxel is concentrated wherever SPARC overexpression exists.11 The result is that more paclitaxel ends up in tumor cells due to albumin nanoparticles.18 Expression profiling of ACC samples in our labo- ratory showed high levels of SPARC messenger RNA expression, prompting further investigation into the potential for nab-paclitaxel as a treatment for ACC. This report details the in vitro and in vivo
From the *Virginia G. Piper Cancer Center, Scottsdale, AZ; Translational Ge- nomics Research Institute, Phoenix, AZ; }Arizona Cancer Center, Tucson, AZ; §Medical College of Wisconsin, Milwaukee, WI; and [Mayo Clinic, Rochester, MN.
Disclosure: Supported by the Advancing Treatments for Adrenocortical Carcinoma (ATAC) Research Fund.
Reprints: Michael J. Demeure, MD, MBA, Translational Genomics Research In- stitute, Scientific Director Endocrine Tumors Center, Virginia G. Piper Cancer Center, 10460 N 92nd St, Suite 200, Scottsdale, AZ 85258. E-mail mdemeure@ tgen.org.
Copyright @ 2011 by Lippincott Williams & Wilkins ISSN: 0003-4932/11/25501-0140
DOI: 10.1097/SLA.0b013e3182402d21
data that support further investigation of nab-paclitaxel as a potential therapeutic agent for ACC.
METHODS
Clinical Samples
Tumor samples used included a total set of 19 ACC flash frozen tumors collected at the Mayo Clinic in Rochester, MN, the Univer- sity Hospital Essen (Essen, Germany), and the University of Calgary (Alberta, Canada). Formalin-fixed paraffin-embedded tissue from an additional 9 ACC (including a matched primary and recurrence), 6 benign adrenocortical adenomas (ACAs), and 1 normal adrenal gland were available for immunohistochemistry staining, for a total of 34 separate tumor samples from 33 patients. The study was con- ducted under Western Institutional Review Board-approved protocol 20051769.
Expression Microarrays and Reverse Transcriptase Quantitative Polymerase Chain Reaction Validation
Tumors were analyzed for messenger RNA expression pro- files using the Affymetrix Human U133A Plus 2 array. Ribonucleic acid was extracted from 100-mg samples of ACC tumors and nor- mal adrenal tissue, amplified and reverse transcribed utilizing the MessageAmp II Biotin Enhanced Kit (Ambion, Inc). Biotin-labeled antisense ribonucleic acid (aRNA) was synthesized according to this standard protocol, followed by purification through provided aRNA Filter Cartridges. Labeled complementary RNA was fragmented and hybridized to arrays following the standard Affymetrix protocol.
Total RNA was reverse transcribed utilizing random hexamer primers and the iScript cDNA Synthesis kit (Bio-Rad Laboratories, Inc). The resulting complementary DNA was amplified on the iQ5 Real-Time PCR Detection System (Bio-Rad Laboratories, Inc) us- ing the iScript RT-PCR Kit with SYBR® green and gene-specific primers designed to span the closest intron-exon junction of the ref- erence sequence to which the probes on the array were designed. For SPARC (GenBank Assession No. NM_003118), the forward and reverse primers were TCT GAC TGA GAA GCA GAA GCT GCG G and CCG AAC TGC CAG TGT ACA GGG AAG A, respectively. For amplification, the following program was employed: a 50℃ pre- heat step for 2 minutes, a 95°℃ heat activation step for 2 minutes, followed by 40 cycles of denaturation at 95℃ for 15 seconds, anneal- ing at 62℃ for 30 seconds, and elongation at 72℃ for 30 seconds. Melting curve analysis was performed to evaluate primer set speci- ficity. Beta-actin was used as the reference gene. Fold difference in cDNA concentration was calculated using the Pfaffl method taking into account reaction efficiencies.19
In Vitro Drug Dose-Response Curves
For in vitro testing of drug efficacy, 2 human adrenocortical cancer cell lines were used. NCI-H295R and SW-13 were obtained from American Type Culture Collection (ATCC), cultured in appro- priate media according to ATCC recommendations, and incubated in 5% CO2 or free gas exchange, respectively. Cells were plated in 0.1-mL medium (containing 2% serum) in 96-well microtiter plates and allowed to grow overnight. On the second day, serial dilutions of the test drugs were added in replicates of 3 to the plates. Cells were then incubated for additional 3 days (72 hours) at 37℃ in a hu- midified incubator. After drug treatment, cells were fixed with 10% trichloroacetic acid for 1 hour at 4℃. After fixation, cells were washed with water, stained with a 0.04% sulfa rhodamine B solution for 1 hour. Cells were then washed with a 1% acetic acid solution. The plates sat at room temperature until dry, and 50 mM Tris/HCl was then added to each well and incubated for 15 minutes. Absorbance at 570 nm was quantified using a plate reader (BioTek, Winooski,
@ 2011 Lippincott Williams & Wilkins
VT). The surviving fraction of cells was determined by dividing the mean absorbance values of the drug-treated samples by the mean absorbance values of the untreated control.
Dose-response curves and IC50 values for cell survival in the presence of the drugs were calculated using Prism4 software (Graph- Pad) using the following equation:
Y = (Top - Bottom)/{1+ 10[(logICso-X)-HillSlope]],
where X is the logarithm of concentration and Y is the percent cell survival. Y starts at the top and goes to bottom with a sigmoid shape as X increases.
In Vivo Mouse Xenograft Studies
For in vivo testing of drug efficacy, xenograft models in IcrTac:ICR-Prkdescid mice were used. Tumor weight and mouse body weight were measured twice weekly. Data collection was concluded with animal morbidity because of tumor burden or when tumor size reached 2 g. The human H295R ACC and SW-13 cell lines were utilized in the creation of mouse xenograft tumors. All procedures were carried out under the institutional guidelines of TGen Drug De- velopment’s Institutional Animal Care and Use Committee (Protocol #06001, approved January 2006). To create xenografts, 1.0 × 107 cells were injected subcutaneously in a solution of 50% media/50% Matrigel into the right flank of the animal. Well-established tumors (~500 mg) were serially passaged by trochar until stable growth was maintained (passage 3). Stable-growing tumor pieces were im- planted 40 days before treatment initiation. When tumors reached a size of 95 mg, cohorts were assigned by random equilibration and compound administration was begun. Tumors were measured using Vernier calipers and tumor weight was calculated using the following formula:
Tumor weight (mg) = (ab2/2),
where “b” is the smallest diameter and “a” is the largest diameter. Thirty-five mice with tumor sizes of 62.5 to 220.5 mg were pair- matched into the 7 groups of 5 mice each by random equilibration (day 1). Body weights were recorded when the mice were pair-matched. In addition, body weights were taken twice weekly thereafter in con- junction with tumor measurements. On day 1, mitotane was diluted with a Corn Oil solution immediately before dosage administration to a working concentration of 30 mg/mL to deliver 300 mg/kg doses at a 0.2-mL fixed dose volume for SW13 and 20 mg/mL for 200 mg/kg dose for H295R. Mitotane was given daily by oral gavage admin- istration. Before administration, nab-paclitaxel was reconstituted by slowly adding 20 mL of 0.9% NaCl. The stock solution was further diluted with 0.9% NaCl solution to a concentration of 3 mg/mL to deliver intravenously a 30 mg/kg dose in a 10 ml/kg dose volume. Paclitaxel was received in clinical stock formulation and was diluted with a 0.9% NaCl solution to a working concentration of 1 mg/mL to deliver 10 mg/kg doses intravenously in a 10 mL/kg dose volume. Nab-paclitaxel and paclitaxel were given on a daily × 5 schedule. When the mean tumor weight of the control group reached 1200 mg, the mice were killed with regulated CO2. At the time of killing, tu- mors were excised and bisected lengthwise. Half of the tumor was snap frozen in RNA later and stored at - 80℃. The other half was fixed in formalin and paraffin embedded.
Mean tumor growth inhibition (TGI) was calculated utilizing the following formula:
TGI = |1 - (X Treated (Final) - X Treated (Day 1)) (¿ Control (Final) - X Control (Day 1) ] × 100%
Tumors that regressed from the Day 1 starting size were removed from the calculations. Individual tumor shrinkage (TS) was calculated
using the formula below for tumors that showed regression relative to Day 1 tumor weight. The mean tumor shrinkage of each group was calculated and reported.
TS = 1 1 (Tumor Weight (Day 1)) (Tumor Weight (Final))
× 100%
All statistical analyses in the xenograft study were performed with GraphPad Prism v4 software.
Immunohistochemistry
The xenograft tumors were fixed in 10% buffered formalin upon harvesting and then processed and paraffin embedded in a stan- dard manner. A xenograft tissue microarray (TMA) was constructed from the treated mice. A manual tissue arrayer (Advanced EDM Au- tomation, Poway, CA) was used to construct the TMAs. All tumor cases were histologically reviewed and representative tumor areas marked on the corresponding donor paraffin blocks. Two 0.6-mm cores from 2 different regions of the tumor were punched and trans- ferred to the recipient block. The xenograft TMA and human tissue whole sections were sectioned at 5-um thickness using water flota- tion for tissue section transfer and dried overnight at room temper- ature. All the slides were dewaxed and rehydrated, and antigen was retrieved online on the BondMax autostainer (Leica Microsystems, Inc, Bannockburn, IL). All slides were subjected to heat-induced epi- tope retrieval using a y citrate-based retrieval solution for 20 minutes. Endogenous peroxidase was blocked. A universal protein block was used and a rodent block M (Biocare Medical, Inc) to block endoge- nous mouse IgG and nonspecific background staining. The TMA was incubated for 25 minutes with an anti-ABCB1 (MDR1) antibody (Lifespan Biosciences, Seattle, WA) at 1:75. Whole sections were incubated at 1:200 for 30 minutes. The TMA was incubated for 30 minutes with an anti-SPARC (goat IgG) antibody (R&D Systems, Inc, Minneapolis, MN) at 2 µg/mL and 7.5 µg/ml for 30 minutes for the whole sections. The TMA sections were visualized using a com- bination of the Bond Polymer Refine Detection kit (Leica) and mouse anti-goat IgG 4plus detection system (Biocare Medical, Inc) using diaminobenzidine chromogen as substrate. The whole sections were visualized using the Bond Polymer Refine Detection kit (Leica). Im- munohistochemistry slides were evaluated for staining intensity (scale of 0-3) in cytoplasmic localization for both SPARC and ABCB1.
Molecular Concept Analysis
The top 1% of genes corresponding to the high-grade ACC samples, as determined by signal-to-noise method, was up- loaded into the Molecular Concept Map utility of Oncomine (www.oncomine.org). The molecular concept map allows for the sys- tematic query of a given set of genes (top 1% high-grade ACC) for overlap similarity with a large compendium of oncology-related gene expression signatures. Fisher exact test was used to measure overlap as implemented in the molecular concept map Web site. Results are visualized in a molecular concept map. Each node represents a partic- ular Oncomine gene expression signature. The diameter of the node is sized relative to the number of genes in the concept. The edges of the graph represent a statistically significant association between the top 1% ACC signature (Fisher exact test P < 0.001).
RESULTS
Clinical Data
Nineteen frozen adrenocortical cancer specimens were avail- able for gene expression analysis, along with 2 normal adrenal glands and 2 normal adrenal gland reference RNA pools. Formalin-fixed
paraffin-embedded tissue was available for an additional 9 ACC (in- cluding a primary and recurrence from the same patient), 6 ACA, and 1 normal adrenal gland. In all, there were tumor specimens from 13 men (10 ACC and 3 ACA) and 20 women (18 ACC and 3 ACA). The average age of the patients at the time of their operation was 50.8 ± 14.9 (µ ± o) years (range, 23-77). Seven of the tumors were nonfunctional, 6 produced the Cushing syndrome, 1 produced aldosterone, and 17 cases were of unknown presentation. For ACC samples, the tumor weight was 623 ± 692 (u ± o) g with a mean tumor diameter of 11.8 ± 5.2 (µ ± o) cm. Benign adrenal adenoma samples had a mean tumor weight of 44 ± 18 g and a mean tumor diameter of 3.7 ± 1.1 cm.
Expression Array Data
The transcriptomes of 19 ACC tumors and 4 normal adrenal glands were assessed on the Affymetrix U133 Plus 2 (Table 1). One gene that was determined to be overexpressed was SPARC, with a mean log2-fold change of 1.56 ± 0.44 (u ± SD). For vali- dation, we performed reverse transcriptase quantitative polymerase chain reaction for SPARC in 6 of the ACC tumors profiled (3 low grade and 3 high grade) and demonstrated a mean of 2.19 ± 1.26- fold overexpression relative to the beta-actin reference. ABCB1, which encodes P-glycoprotein, had a mean log2-fold change of 0.37 ± 0.39 (µ ± SD) on the microarray. However, as detailed in Table 2, there was variability between tumors, a subset of tumors that overexpressed, underexpressed, or had normal expression of SPARC or P-glycoprotein.
Immunohistochemistry
Eight of 11 ACC samples from 9 patients stained positive or weakly positive for cytoplasmic/membranous SPARC (Table 3). Five of the 6 ACA samples showed cytoplasmic/membranous staining. One ACA had nuclear localization of SPARC. The normal adrenal gland had weak staining for SPARC. P-glycoprotein was expressed in 10 of 11 ACC and 5 of 6 ACA. However, 6 of 11 ACC had staining intensities that were less than observed in the normal adrenal gland. There was no correlation between SPARC expression and P- glycoprotein expression.
In Vitro and In Vivo Data
There was no significant difference in drug effect on the growth of the 2 ACC cell lines between paclitaxel and nab-paclitaxel in vitro (Fig. 1). In contrast, in vivo experiments in murine xenografts looking at the effect of nab-paclitaxel compared with the effect of mitotane and paclitaxel showed a difference between drugs in both cell lines. There was, in the H295R cells, a much greater growth inhibition of tumor in nab-paclitaxel-treated mice when compared with those treated by standard paclitaxel (Fig. 2A). In the SW-13 cell line, pacli- taxel and nab-paclitaxel showed similar tumor growth inhibition, and
| Gene | Affymetrix Probe ID | GenBank Sequence | Average Log Fold Change | SD |
|---|---|---|---|---|
| SPARC | 200665_s_at | NM_003118 | 1.87 | 0.79 |
| 212667_at | 1.42 | 0.97 | ||
| ABCB1 | 209993_at | NM_000927 | 0.37 | 0.39 |
*Expression of ABCB1 gene for the multidrug resistance P-glycoprotein shows that on average across all ACC samples, transcript levels are decreased.
both were more effective than mitotane (Fig. 2B). The difference in drug response between the 2 cell lines can be explained by increased expression of SPARC as detected by immunohistochemistry (IHC) on the xenograft specimens noted in H295R cells compared with SW- 13 (Fig. 3). Increased SPARC expression then is consistent with the observation of an increased effect of nab-paclitaxel compared with paclitaxel seen in H295R but not seen in SW-13.
| Gene | SPARC | SPARC | ABCB1 |
|---|---|---|---|
| Probe | 200665_s_at | 212667_at | 209993_at |
| Reference sequence | NM_003118 | NM_000927 | |
| Tumor sample ID | Fold change relative to normal adrenal gland | ||
| ACC4 | 1.72 | 0.87 | 0.48 |
| ACC5 | 3.13 | 1.69 | 0.40 |
| ACC8 | 0.67 | 0.46 | 0.02 |
| ACC9 | 1.39 | 0.85 | 0.12 |
| ACC10 | 0.14 | 0.05 | 0.01 |
| ACC11 | 1.19 | 0.69 | 0.12 |
| ACC12 | 2.82 | 1.81 | 1.07 |
| ACC13 | 1.61 | 0.90 | 0.29 |
| ACC17 | 2.65 | 2.82 | 0.07 |
| ACC19 | 1.94 | 0.95 | 0.15 |
| ACC21 | 2.45 | 1.55 | 0.64 |
| ACC22 | 1.24 | 1.13 | 0.03 |
| ACC23 | 1.71 | 1.82 | 0.21 |
| ACC26 | 2.65 | 0.87 | 0.58 |
| ACC27 | 2.34 | 1.35 | 0.09 |
| ACC28 | 1.77 | 1.08 | 0.85 |
| MPI1 | 2.63 | 2.63 | 0.07 |
| MPI2 | 2.69 | 3.11 | 0.03 |
| MPI3 | 1.12 | 1.42 | 0.39 |
Knowledge Mining
The ACC gene expression signature representing the top 1% of genes associated with high-grade tumors was analyzed to identify similar oncogenic signatures. The top 1% genes were uploaded into the Oncomine (www.oncomine.org) database and molecular concept analysis was performed. Notably, 13 of 20 of the top signatures were of breast cancer origin (Fig. 4).
DISCUSSION
Targeted rational therapy is particularly needed in efforts to find new treatments for rare cancers where traditional methods of drug discovery encounter serious drawbacks and have had few successful results. For some rare tumor types, there is little or no evidence-based standard therapy. In other cases, a tumor may be so rare that it be- comes difficult to accrue sufficient numbers of patients in to a clinical study with adequate statistical power to address efficacy. Thus, when
120-
# SW-13/Paclitaxel
NCI-H295R/Paclitaxel
100-
· SW-13/nab-Paclitaxel
· NCI-H295R/nab-Paclitaxel
% Cell Survival
80-
60-
40-
20-
0
10-11
10-10
10-9
10-8
10-7
10-6
10-5
[Drug]/M
| Sample | Tissue Type | ABCB1 | SPARC | ||||
|---|---|---|---|---|---|---|---|
| Nuclear | Cytoplasmic | Composite | Nuclear | Cytoplasmic | Composite | ||
| ACC 34 | ACC | 0 | 1 | 1 | 0 | 0 | 0 |
| ACC 93 | ACC | 0 | 2 | 2 | 0 | 2 | 2 |
| ACC 81 | ACC | 0 | 0 | 0 | 0 | 2 | 2 |
| ACC 73 | ACC | 0 | 2 | 2 | 0 | 1 | 1 |
| ACC 61 | ACC | 0 | 1 | 1 | 0 | 2 | 2 |
| ACC 69 | ACA | 0 | 1 | 1 | 2 | 0 | 2 |
| ACC 55 | ACA | 0 | 1 | 1 | 0 | 1 | 1 |
| MPI1 | ACC | 0 | 2 | 2 | 0 | 0 | 0 |
| ACC 86 | Norm | 0 | 2 | 2 | 0 | 1 | 1 |
| ACC 62 | ACA | 0 | 2 | 2 | 0 | 2 | 2 |
| ACC 65 | ACA | 0 | 2 | 2 | 0 | 1 | 1 |
| ACC 68 | ACA | 0 | 0 | 0 | 0 | 0 | 0 |
| ACC 59A | ACC | 0 | 1 | 1 | 0 | 1 | 1 |
| ACC 59B | ACC | 0 | 3 | 3 | 0 | 1 | 1 |
| ACC 85 | ACC | 0 | 1 | 1 | 0 | 0 | 0 |
| ACC 110 | ACC | 0 | 1 | 1 | 0 | 1 | 1 |
| ACC 113 | ACC | 0 | 3 | 3 | 0 | 2 | 2 |
| ACC 67 | ACA | 0 | 2 | 2 | 0 | 1 | 1 |
0 indicates negative; 1, weak positive; 2, positive; 3, strong positive.
A
H295R
2000
- Control
Tumor weight (mg)
Mitotane (150 mg/kg)
1500-
— Paclitaxel (10mg/kg)
- nab-Paclitaxel (30 mg/kg)
1000
500
0
0
5
10
15
20
25
30
35
Days
SW-13
B
2000
Control
Tumor weight (mg)
*- Mitotane (300 mg/kg)
1500
— Paclitaxel (10mg/kg)
+ nab-Paclitaxel (30 mg/kg)
1000-
500-
0
0
5
10
15
20
25
30
35
40
45
Days
developing therapies for rare cancers, traditional approaches culmi- nating in large phase III clinical trials of a novel treatment against a supposed standard treatment may not even be possible. Furthermore, the potential financial reward may not justify the cost of traditional drug development and large-scale clinical studies. The Food and Drug Administration has recognized these difficulties and has offered in- centives to encourage the development of new agents for rare diseases in its Orphan Diseases program. One clear benefit that the Food and Drug Administration offers to manufacturers whose drugs are ap- proved through this program is extended patent protections.
A novel means of finding new treatments for patients with rare cancers is an imperative. Our study shows a possible new clinical indication for an already Food and Drug Administration-approved agent, nab-paclitaxel. We used a genomics-based approach to identify SPARC as a potential new therapeutic target in ACC. We show that SPARC is expressed in ACC at various levels, by both messenger RNA and IHC. In a xenograft model of ACC, both paclitaxel and nab-paclitaxel demonstrate greater growth inhibition compared with mitotane, and the expression of SPARC is correlated with differential response to nab-paclitaxel as compared with paclitaxel. The novel technologies used in this project include genomic technologies and bioinformatics knowledge mining. These approaches demonstrated that there is significant overlap with the top 1% high-grade ACC signature with a number of advanced breast cancer signatures. This is notable in that the current indication approved, in the United States, for nab-paclitaxel is in refractory breast cancer. In a study of 454 women with metastatic breast cancer,13 treatment with nab-paclitaxel was associated with a greater response rate (33% vs 19%, P < 0.001) and a longer time to tumor progression (21.9 vs 16.1 weeks, P = 0.029) than
SW-13
H295R
treatment with Cremophor-based paclitaxel. Grade 4 neutropenia was more common in patients treated with Cremophor-based paclitaxel than in patients treated with nab-paclitaxel (22% vs 9%, P < 0.001), but Grade 3 neuropathy, although manageable, was more common among patients treated with nab-paclitaxel. Furthermore, although no difference in survival between the agents was seen when used as a first-line therapeutic regimen, in patients who received nab-paclitaxel as a second-line treatment or greater, survival time was longer than in those who received standard paclitaxel (56.4 vs 46.7 weeks, hazard ratio = 0.73, P = 0.024). Further work is needed to investigate if genes within the overlapping signatures between ACC and breast cancer indicate sensitivity to the nab-paclitaxel or if any particular gene signature can serve as a potential marker of response.
It is not yet clear whether nab-paclitaxel is affected by P-glycoprotein multidrug resistance to the extent that it af- fects paclitaxel.19,20 P-glycoprotein is an adenosine triphosphate- dependent efflux pump that reduces chemotherapy drug tumor cell accumulation and mediates resistance to the drug. Standard drugs used for patients with ACC in the Italian regimen including doxoru- bicin and etoposide are known substrates for P-glycoprotein.21 In our ACC study, we found great variability in P-glycoprotein levels by expression array analysis and by IHC suggesting that patients whose tumors have high level of SPARC and low levels of P-glycoprotein could be candidates for treatment with nab-palcitaxel. Furthermore, our murine xenograft studies did not show induction of P-glycoprotein over time during treatment with nab-paclitaxel.
A limitation of our study is the relatively small number of tumor samples studied. As ACC is a rare disease, access to frozen tumor samples, in particular, is difficult. In this case, 19 cancers is a relatively large sample size for this type of analysis if one compares this project to others previously published. For rare diseases, it is important for busy clinical centers to establish tumor repositories with sample annotation with clinically relevant data such as response to treatment. A phase II study of a novel treatment in ACC would need a multicenter study for adequate accrual. For a randomized phase III trial comparing treatments in ACC, one would likely need to rely on a multinational study as is the case with the First International Randomized trial in locally advanced and Metastatic Adrenocortical Carcinoma Treatment trial that has just completed accrual.
@ 2011 Lippincott Williams & Wilkins
Breast carcinoma elston grade - Top 5% overexpressed in 3 (Miller)
Ovarian carcinoma grade-Top 1% overexpressed in 3 (Bowtell)
Breast carcinoma grade-Top 5% overexpressed in 3 (Bittner)
Breast carcinoma genomic risk (Veridex signature)-Top 5% overexpressed in high risk (Desmedt)
Lung carcinoma grade-Top 5% overexpressed in 4 (Bittner)
Hepatocellular carcinoma vascular invasion-Top 5% overexpressed in microscopic, macroscopic (Wurmbach)
Cancer type - Top 5% underexpressed in prostate carcinoma (Bittner)
Breast carcinoma elston grade - Top 5% overexpressed in 3 (Ivshina)
Breast carcinoma grade - Top 1% overexpressed in 3 (Sotiriou)
Breast carcinoma treatment status-Top 1% underexpressed in postletrozole treatment (Miler)
Breast carcinoma grade-Top 1% overexpressed in 3 (Van)
Breast carcinoma grade-Top 1% overexpressed in 3 (Ma)
Breast carcinoma diferentiation-Top 1% overexpressed in poor (Desmedt)
Breast carcinoma p53 mutation status-Top 5% overexpressed in mutant (Miler)
Breast carcinoma grade-Top 1% overexpressed in 3 (Ginestier)
Hepatocellular carcinoma differentiation- Top 5% overexpressed in poor (Wurmbach)
Breast carcinoma type-Top 5% overexpressed in ductal breast carcinoma (Bittner)
Breast carcinoma p53 mutation status-Top 5% over expressed in mutant (Ivshina)
ACC_HighGrade_Top1
Vulva type-Top 5% overexpressed in vulvar intraepithelial neoplasia (Santegoets)
Cancer type-Top 10% overexpressed in cervical carcinoma (Bittner)
In summary, this article presents data in support of conducting further investigations of the use of nab-paclitaxel in patients with ACC. A phase II trial of nab-paclitaxel in patients with ACC would be welcome for this rare disease where there is a paucity of proven treatment options. We used novel methods including expression pro- filing of tumors and knowledge mining of the literature to iden- tify SPARC as a promising therapeutic target. Targeted therapy with nab-paclitaxel should be investigated particularly in patients whose ACC express high levels of SPARC and low levels of pglycopro- tein/multidrug resistance protein 1.
ACKNOWLEDGMENTS
The authors thank Dr. Yu Zhao and April Watanabe for their contributions to the article. They also thank Dr Jung-Han Kim for his technical assistance. This work was supported by the Advancing Treatments for Adrenocortical Carcinoma Research Fund.
REFERENCES
1. Icard P, Chapuis Y, Andreassian B, et al. Adrenocortical carcinoma in sur- gically treated patients: a retrospective study on 156 cases by the French Association of Endocrine Surgery. Surgery. 1992; 112:972-979; discussion 979-980.
2. Demeure MJ, Somberg LB. Functioning and nonfunctioning adrenocortical carcinoma: clinical presentation and therapeutic strategies. Surg Oncol Clin N Am. 1998;7:791-805.
3. Bergenstal DM, Lipsett M, Moy RN, et al. Regression of adrenal cancer and suppression of adrenal function in man by o,p’-DDD. Trans Am Phys. 1959;72:341-350.
4. Decker RA, Elson P, Hogan TF, et al. Eastern Cooperative Oncology Group study 1879: mitotane and adriamycin in patients with advanced adrenocortical carcinoma. Surgery. 1991;110:1006-1013.
5. Hahner S, Fassnacht M. Mitotane for adrenocortical carcinoma treatment. Curr Opin Investig Drugs. 2005;6:386-394.
6. Berruti A, Terzolo M, Pia A, et al. Mitotane associated with etoposide, dox- orubicin, and cisplatin in the treatment of advanced adrenocortical carcinoma. Italian Group for the Study of Adrenal Cancer. Cancer. 1998;83:2194-2200.
7. Flynn SD, Murren JR, Kirby WM, et al. P-glycoprotein expression and mul- tidrug resistance in adrenocortical carcinoma. Surgery. 1992;112:981-986.
8. Abraham J, Bakke S, Rutt A, et al. A phase II trial of combination chemotherapy and surgical resection for the treatment of metastatic adrenocortical carcinoma: continuous infusion doxorubicin, vincristine, and etoposide with daily mitotane as a P-glycoprotein antagonist. Cancer. 2002;94:2333-2343.
9. Rowinsky EK, Cazenave LA, Donehower RC. Taxol: a novel investigational antimicrotubule agent. J Natl Cancer Inst. 1990;82:1247-1259.
10. Fallo F, Pilon C, Barzon L, et al. Paclitaxel is an effective antiproliferative agent on the human NCI-H295 adrenocortical carcinoma cell line. Chemotherapy. 1998;44:129-134.
11. Sparreboom A, Scripture CD, Trieu V, et al. Comparative preclinical and clinical pharmacokinetics of a cremophor-free, nanoparticle albumin-bound paclitaxel (ABI-007) and paclitaxel formulated in Cremophor (Taxol). Clin Cancer Res. 2005;11:4136-4143.
12. Desai N, Trieu V, Yao Z, et al. Increased antitumor activity, intratumor pacli- taxel concentrations, and endothelial cell transport of cremophor-free, albumin- bound paclitaxel, ABI-007, compared with cremophor-based paclitaxel. Clin Cancer Res. 2006;12:1317-1324.
13. Gradishar WJ, Tjulandin S, Davidson N, et al. Phase III trial of nanoparti- cle albumin-bound paclitaxel compared with polyethylated castor oil-based paclitaxel in women with breast cancer. J Clin Oncol. 2005;23:7794-7803.
14. Clark CJ, Sage EH. A prototypic matricellular protein in the tumor microenvironment-where there’s SPARC, there’s fire. J Cell Biochem. 2008;104:721-732.
15. Reed MJ, Sage EH. SPARC and the extracellular matrix: implications for cancer and wound repair. Curr Top Microbiol Immunol. 1996;213:81-94.
16. Thomas R, True LD, Bassuk JA, et al. Differential expression of os- teonectin/SPARC during human prostate cancer progression. Clin Cancer Res. 2000;6:1140-1149.
17. Wang CS, Lin KH, Chen SL, et al. Overexpression of SPARC gene in hu- man gastric carcinoma and its clinic-pathologic significance. Br J Cancer. 2004;91:1924-1930.
@ 2011 Lippincott Williams & Wilkins
18. Nyman DW, Campbell KJ, Hersh E, et al. Phase I and pharmacokinetics trial of ABI-007, a novel nanoparticle formulation of paclitaxel in patients with advanced nonhematologic malignancies. J Clin Oncol. 2005;23:7785-7593.
19. Hennenfent KL, Govindan R. Novel formulations of taxanes: a review. Old wine in a new bottle? Ann Oncol. 2006;17:735-749.
20. Jang SH, Wientjes MG, Au JL. Kinetics of P-glycoprotein-mediated efflux of paclitaxel. J Pharmacol Exp Ther. 2001;298:1236-1242.
21. Berruti A, Terzolo M, Sperone P, et al. Etoposide, doxorubicin and cisplatin plus mitotane in the treatment of advanced adrenocortical carcinoma: a large prospective phase II trial. Endocr Relat Cancer. 2005;12:657-666.