| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Trials |
Departments of Otolaryngology [Q. Z., T. J. S., J. R. G.] and Medicine [D. L. T.], University of Pittsburgh School of Medicine and the University of Pittsburgh Cancer Institute [W. E. G.], Pittsburgh, Pennsylvania 15213, and the Department of Medicine, University of Michigan School of Medicine, Michigan [D. C. S.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and epidermal growth factor
receptor (EGFR) levels in head and neck squamous cell carcinoma by
decreasing the transcription rate of these two genes. Previous
reports suggest that RA receptor (RAR)-ß levels are down-modulated in
head and neck cancer and are restored by RA therapy. Cellular
RA-binding protein (CRABP)-II is up-regulated by RA and appears to
modulate intracellular RA metabolism. In conjunction with a Phase I
clinical trial, total intact RNA was extracted from oral cavity mucosa
biopsied from 17 patients with advanced malignancies, before and after
treatment with a 4-week course of liarozole. To analyze these limited
quantities of total RNA (as little as 0.6 µg/sample), a quantitative
reverse transcription-PCR assay was developed using delayed dropping of
the 5' ß-actin primer to amplify the highly abundant ß-actin gene
as an internal control. We used this method to determine the expression
levels of TGF-, EGFR, RAR-ß, and CRABP-II before and after
treatment. There was a trend toward elevation of RAR-ß levels in oral
mucosa after liarozole therapy (P = 0.107), whereas
TGF-, EGFR, and CRABP-II were not modulated by systemic liarozole
treatment. These results suggest that liarozole may up-regulate RAR-ß
in tissues from cancer patients and that expression levels of potential
intermediate biomarkers may be determined in small tissue biopsies
using a quantitative reverse transcription-PCR assay. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Rubin Grandis and Tweardy (13)
and Rubin Grandis et
al. (14)
reported previously that TGF- and EGFR
mRNA and protein are up-regulated both in histologically normal mucosa
several centimeters away from the tumor and in tumors from patients
with head and neck SCC when compared with levels in control normal
mucosa from individuals without cancer, suggesting that this
up-regulation is an early event in head and neck carcinogenesis. This
up-regulation appears to be due to activated gene transcription (as
opposed to increased DNA or prolongation of mRNA half-life) and can be
down-modulated by RA to approach the transcription rate of normal
mucosal epithelial cells (15)
. The potential utility of
monitoring TGF- and EGFR expression levels in chemoprevention trials
is further supported by our demonstration of increased TGF- and EGFR
protein in premalignant dysplastic lesions compared with control normal
mucosa (16)
.
The use of retinoids in the clinical setting has been limited by their toxicity, which includes skin, ocular, and muscle abnormalities; pseudo-tumor cerebri; hepatotoxicity; and hypertriglyceridemia (17) . Liarozole is an inhibitor of RA metabolism that is thought to exert its biological effects through elevation of endogenous RA, with a reduced side effect profile compared with synthetic retinoids (18) . The ability to predict response to therapy by assessing modulation of target gene expression levels would enable the clinician to determine the likely efficacy of treatment before definitive clinical outcome. To test the hypothesis that putative intermediate biomarker gene expression levels could be determined in surrogate tissue using a quantitative RT-PCR assay, we obtained oral mucosa biopsies before and immediately after a 4-week course of liarozole, in conjunction with a multi-institutional Phase I clinical trial. Measurement of gene expression levels in small tissue biopsies was accomplished using a quantitative RT-PCR assay.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RNA Preparation.
Total RNA was extracted from tissues that were dissected using the
RNeasy mini-kit (Qiagen, Santa Clarita, CA). Cell lysates were
homogenized by passage through a 23-gauge needle repeatedly followed by
spinning through a QIAshredder (Qiagen). RNA concentrations were
determined by absorbance at 260 nm UV absorbance using a
spectrophotometer (Pharmacia Biotech, Piscataway, NJ). The mean total
RNA concentration from pretreatment biopsies was 8.15 µg (range,
0.6–54.52 µg), and the mean total RNA concentration from
posttreatment biopsies was 7.51 µg (range, 0.66–25.98 µg).
RT-PCR.
RT-PCR reactions were performed using two pairs of primers in each
reaction with the SuperScript one-step RT-PCR system (Life
Technologies, Inc., Gaithersburg, MD). One pair of primers was specific
for the target intermediate biomarker gene, and the other pair was
specific for ß-actin to be used as an internal control for
quantitation (see Table 1
for primer
sequences and PCR product lengths). Because all of the target genes
were less abundant than ß-actin, and coamplification of mRNAs present
at different levels might result in competitive interference, we
preamplified the target genes with several cycles of PCR after the RT
reaction before "dropping in" the sense ß-actin primer to
coamplify both genes for the remaining PCR cycles. The RT-PCR reaction
volume was scaled down to 20 µl. A total of 0.01 µg of total RNA
was used as a template in the reaction mixture with 1x RT-PCR buffer,
0.5 µl of 20 µM sense and antisense primers of the
target gene, 0.5 µl of 20 µM ß-actin antisense
primer, 0.3 µl of [-32P]dCTP (6000
Ci/mmol; 20 mCi/ml; New England Nuclear, Boston MA), and 0.5 µl of
SuperscriptII RT/Taq mix. The RT reaction was performed for 30 min at
50°C and ended by incubating the reaction at 94°C for 2 min. PCR
conditions for TGF- and CRABP-II were 94°C for 30 s, 60°C
for 30 s, and 72°C for 30 s. PCR conditions for EGFR and
RAR-ß were 94°C for 30 s, 60°C for 30 s, and 72°C for
45 s. After preamplification of the target gene (TGF-, 6
cycles; EGFR and CRABP-II, 8 cycles; RAR-ß, 14 cycles), 5 µl of a
solution containing 1x RT-PCR buffer and 0.5 µl of 20
µM ß-actin sense primer were added to each reaction,
followed by 18 additional cycles.
|
-32P]dCTP in the PCR products, which were
in proportion to the density of each electrophoresed band) were scanned
using the computerized PhosphorImager and analyzed by ImageQuant
software (Molecular Dynamics). The ratio of the target gene:ß-actin
was used to designate gene expression levels. Designated gene
expression levels in the mucosa before liarozole treatment were
compared with expression levels after treatment.
Statistics.
A paired t test was used to compare total RNA in pre- and
posttreatment biopsies. Treatment effects were calculated as the
difference between pretreatment and posttreatment values. An exact
signed rank test was then conducted to test for significant differences
in any of the four intermediate biomarkers. The effect of liarozole
dose on treatment effects was examined with the Kruskal-Wallis test and
Jonckheere’s test for trend using the exact permutation distribution
of each test statistic.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, EGFR, and CRABP-II in patient 17; and CRABP-II in
patient 16). There was no difference in the total amount of RNA
isolated from pre- and posttreatment biopsies (P =
0.591; data not shown).
Quantitative RT-PCR Conditions.
A series of preliminary experiments were performed to establish and
optimize the quantitative conditions by using total RNA from unpaired
tissue samples. These studies included determination of the exponential
phase for each PCR amplification, optimization of the primer dropping
time point(s), and examination of the quantifiable range
correlated with total RNA input. For the exponential phase, using
different amounts of total RNA as a template, PCR amplification curves
for each target gene and ß-actin were generated. When 0.1 µg of
total RNA was used in the one-step RT-PCR system, gene amplification
remained in the exponential phase if the cycle number was less than 30
for TGF-, EGFR, and CRABP-II; less than 35 for RAR-ß; and less
than 20 for ß-actin (data not shown). These cycle numbers resulted in
a linear relationship between the amount of mRNA input and PCR product.
Using lower amounts of starting RNA widened the range of the
exponential phase. Therefore, by using less than 0.1 µg of total RNA
as a template to amplify TGF- and less than 30 PCR cycles, we could
insure that gene amplification was in the linear phase, thus avoiding a
"plateau" in detecting the potential modulation of gene expression
levels. Using ß-actin as an internal control, these studies
determined that the amplification cycle number should be less than or
equal to 20.
Primer Dropping Time Point.
To remain in the linear phase, we systematically examined different PCR
cycle numbers for each target gene, alone and in combination with
ß-actin. Using this method, the ideal primer dropping point was
determined for each gene [TGF-, 6 cycles; EGFR and CRABP-II, 8
cycles; RAR-ß, 14 cycles (Table 2)
].
All target genes were coamplified with ß-actin for 18 cycles. This
primer dropping method minimized the competition for PCR substrates
between target and control genes so that both gene products could reach
detectable and comparable levels (Ref. 19
; Fig. 1
).
|
|
, target gene determinations were most
accurate when the sample was less than 0.08 µg of total RNA. The RNA
input range for all target genes studied was 0.01–0.08 µg. Using
nonradioactive methods, PCR products were difficult to detect when the
total RNA input was below 0.005 µg (Fig. 2)
. This experiment was
repeated twice for each gene, with similar results. To increase the
sensitivity of the detection method, we introduced
[-32P]dCTP into the RT-PCR reactions as part
of the DNA synthesis substrate. Designated expression levels of the
target and control genes were then determined using a
PhosphorImager. The variability of the established quantitative
RT-PCR method was also evaluated (Fig. 3)
. Because the ratio of target
gene:ß-actin was used to designate expression levels, we measured the
radioactive strength of the PCR products and calculated this ratio in
each reaction using different amounts of template, and then we
determined the variability range. The largest variation was found to be
18.3% greater that the average (0.651). This method allowed us to
detect and measure the final products when the total RNA was diluted
from 0.08 to 0.00125 µg. In this range, the radioactive signal
strength was in proportion to the RNA template input. Using this
technique, we were able to measure up to a 64-fold change in target
gene expression levels. This experiment was repeated three times with a
high degree of consistency in the results calculated.
|
|
. Target gene quantitations obtained
using the paired patient samples were all performed in the
quantifiable range by monitoring the quantitation control.
|
and 6)
.
Experiments were repeated twice with reproducible results. There was a
trend toward elevation of RAR-ß levels after therapy that did not
achieve statistical significance (Fig. 7)
.
|
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Rubin Grandis et al. (15)
reported previously
that TGF- and EGFR gene transcription rates are down-modulated by RA
in vitro. Others have shown that RAR-ß levels are restored
by RA therapy (5)
and that CRABP-II modulates
intracellular RA metabolism and is up-regulated by RA
(22)
. In conjunction with a multi-institutional Phase I
clinical trial, we biopsied oral mucosa before and after a 4-week
course of liarozole therapy to determine the modulation of intermediate
biomarker gene expression levels in surrogate tissue. To accomplish
this, we developed a quantitative RT-PCR assay using a primer dropping
method to measure the expression level of target genes relative to a
control (e.g., housekeeping) gene (e.g.,
ß-actin). Because ß-actin is more abundant than the target gene(s)
examined and because the total RNA obtained from biopsies was often
quite low, it was necessary to devise standard curves for each gene
(primer pair) to determine the linear expression level range relative
to RNA input. Thus, the primers for each target gene required
incubation in the PCR reaction for a precisely determined number of
cycles before the addition of the ß-actin primer pairs.
The results of a RT-PCR-based quantitation are effected by each step, including quantitation of starting material (total RNA), cDNA synthesis (reverse transcription), and PCR reactions. Several quantitative RT-PCR methods have been reported previously, including variations of the primer dropping technique used in this study (18) . In the widely used competitive RT-PCR quantitation method, a known amount of synthesized DNA is used as an internal control for PCR amplification (23 , 24) . However, errors in RNA measurement and template loading and varying efficiencies in different RT reactions may interfere with the results. In some cases, a plasmid construct of a housekeeping gene, such as ß-actin, has been used for competitive RT-PCR purposes (25) . An extra set of PCR reactions using this control aids in calibration of the cDNA used in subsequent PCR quantitation of the target gene. This method minimizes the effects caused by errors in total RNA quantitation and unequal synthesis of cDNA, but inaccuracies in the addition of the PCR template can still affect the results. We used endogenous ß-actin mRNA as an internal control in our PCR quantitation. This method avoided errors by performing sample manipulation, RT, and PCR reactions in the presence of the control. A primer dropping technique controls for differential amplification of different genes, uses two pairs of primers, and minimizes competition for the substrate (19) . Quantitative results can best be achieved with one PCR reaction using this assay instead of a series of reactions using different combinations of template and competitor (control). This one-step method is particularly useful in translational studies in which the amount of patient sample may be extremely limited.
Our results suggest that expression levels of TGF-, EGFR, RAR-ß,
and CRABP-II were not significantly altered in surrogate mucosa,
although there was a trend toward elevation of RAR-ß after liarozole
treatment. There are several possible explanations for the lack of
modulation of target gene expression by liarozole, including:
(a) lack of efficacy of the compound, specifically with
respect to increasing endogenous RA levels (a variable that was not
examined in this study); (b) inappropriate selection of the
target genes; (c) use of surrogate tissue (in contrast to
tumor tissue); and (d) inability to detect statistical
significance due to the relatively small number of patients
(e.g., with RAR-ß and possibly CRABP-II). The
establishment of intermediate end points in cancer treatment requires
that the method of determination be amenable to small biopsy samples.
In addition, tumors are often not readily accessible for biopsy, thus
restricting the analysis to surrogate tissue. Methods that are suitable
for small quantities of RNA should be considered and devised for use in
such studies.
|
FOOTNOTES |
|---|
1 Supported in part by the Janssen Research
Foundation, the Eye and Ear Foundation, the Pittsburgh Foundation, the
John R. McCune Charitable Trust Foundation, and the Mary Hillman
Jennings Foundation. 
2 To whom requests for reprints should be
addressed. Present address: The Eye & Ear Institute, Suite 500, 200
Lothrop Street, Pittsburgh, PA 15213. Phone: (412) 647-5280; Fax: (412)
647-2080.
3 The abbreviations used are: SCC, squamous cell
carcinoma; RA, retinoic acid; TGF, transforming growth factor; EGFR,
epidermal growth factor receptor; RAR, RA receptor; CRABP, cellular
RA-binding protein; RT, reverse transcription; TBE, Tris-borate EDTA.
Received 11/29/99; revised 1/10/00; accepted 2/24/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and EGFR in head and neck cancer. J. Cell. Biochem., 17F(Suppl.): 188-191, 1993.[CrossRef]
and epidermal growth factor receptor in patients with squamous cell carcinoma of the head and neck. Cancer (Phila.), 78: 1284-1292, 1996.[CrossRef][Medline]
and EGFR in head and neck cancer cell lines. Nat. Med., 2: 237-240, 1996.[CrossRef][Medline]
and epidermal growth factor receptor protein expression in progression of premalignant lesions to head and neck squamous cell carcinoma. Clin. Cancer Res., 4: 13-20, 1998.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  R. L. Ferris, L. Xi, S. Raja, J. L. Hunt, J. Wang, W. E. Gooding, L. Kelly, J. Ching, J. D. Luketich, and T. E. Godfrey Molecular Staging of Cervical Lymph Nodes in Squamous Cell Carcinoma of the Head and Neck Cancer Res., March 15, 2005; 65(6): 2147 - 2156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Xi and J.R. Grandis Gene Therapy for the Treatment of Oral Squamous Cell Carcinoma Journal of Dental Research, January 1, 2003; 82(1): 11 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q.-T. Le, P. D. Sutphin, S. Raychaudhuri, S. C. T. Yu, D. J. Terris, H. S. Lin, B. Lum, H. A Pinto, A. C. Koong, and A. J. Giaccia Identification of Osteopontin as a Prognostic Plasma Marker for Head and Neck Squamous Cell Carcinomas Clin. Cancer Res., January 1, 2003; 9(1): 59 - 67. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. N. Lango, K. F. Dyer, V. W. Y. Lui, W. E. Gooding, C. Gubish, J. M. Siegfried, and J. R. Grandis Gastrin-Releasing Peptide Receptor-Mediated Autocrine Growth in Squamous Cell Carcinoma of the Head and Neck J Natl Cancer Inst, March 6, 2002; 94(5): 375 - 383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sudbo, W. Kildal, A. C. Johannessen, H. S. Koppang, A. Sudbo, H. E. Danielsen, B. Risberg, and A. Reith Gross Genomic Aberrations in Precancers: Clinical Implications of a Long-Term Follow-Up Study in Oral Erythroplakias J. Clin. Oncol., January 15, 2002; 20(2): 456 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. I. Song, M. N. Lango, J. D. Hwang, S. D. Drenning, Q. Zeng, W. W. Lamph, and J. R. Grandis Abrogation of Transforming Growth Factor-{alpha}/Epidermal Growth Factor Receptor Autocrine Signaling by an RXR-selective Retinoid (LGD1069, Targretin) in Head and Neck Cancer Cell Lines Cancer Res., August 1, 2001; 61(15): 5919 - 5925. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lotan Are We Ready to Use Surrogate End Points and Surrogate Tissues to Evaluate Response to Chemopreventive and Therapeutic Intervention? Clin. Cancer Res., June 1, 2000; 6(6): 2126 - 2128. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
