Cancer Letters 229 (2005) 33–41
`
`www.elsevier.com/locate/canlet
`
`Non-invasive detection of colorectal tumours by the combined
`application of molecular diagnosis and the faecal occult blood test
`
`Nadine Kutznera, Ingrid Hoffmanna, Christina Linkea, Thomas Thienela,
`Marco Grzegorczykb, Wolfgang Urferb, Dirk Martinc, Gu¨nther Winded, Thilo Traskae,
`Gerd Hohlbache, Klaus-Michael Mu¨llerf, Cornelius Kuhnenf, Oliver Mu¨llera,*
`
`aMax-Planck-Institut fu¨r molekulare Physiologie, Otto-Hahn-Strabe 11, 44227 Dortmund, Germany
`bUniversita¨t Dortmund, 44221 Dortmund, Germany
`cKatholisches Krankenhaus, 44379 Dortmund, Germany
`dKlinik fu¨r Chirurgie, Klinikum Kreis Herford, 32045 Herford, Germany
`eKlinik fu¨r Chirurgie Marienhospital, 44625 Herne, Germany
`fBerufsgenossenschaftlichen Kliniken Bergmannsheil, 44789 Bochum, Germany
`
`Received 16 November 2004; received in revised form 10 December 2004; accepted 12 December 2004
`
`Abstract
`
`The treatment of early-stage tumours decreases the overall mortality of colorectal tumour patients. In this retrospective study
`we determined the sensitivity and the specificity of the faecal occult blood test (FOBT) and the molecular diagnosis (MD). We
`analysed 57 stool samples from patients with colorectal carcinomas for the presence of occult blood using a standard FOBT and
`for alterations in the three different tumour relevant markers APC, BAT26 and L-DNA. Stool samples from 44 control donors
`were analysed to determine the specificity of the applied methods. Twenty-nine (51%; 95% confidence interval (CI): 38–63%)
`stool samples of the cancer patients gave positive FOBT results. Thirty-seven (65%; CI: 52–76%) samples showed alterations in
`at least one DNA marker. Sixteen (28%) samples were positive only in the FOBT, and 24 (42%) samples showed a positive
`result exclusively in MD. The combined application of both methods resulted in a sensitivity of 93% (CI: 83–97%) and an
`overall specificity of 89% (CI: 76–95%). The combined application of FOBT and MD resulted in an overall sensitivity, which
`could not be achieved by any of the methods alone and which is in the range of invasive diagnostic methods.
`q 2005 Elsevier Ireland Ltd. All rights reserved.
`
`Keywords: Non-invasive colorectal cancer diagnosis; Faecal DNA analysis; Molecular diagnosis; Faecal occult blood test
`
`* Corresponding author. Tel.: C49 231 133 2158; fax: C49 231
`133 2199.
`E-mail
`(O. Mu¨ller).
`
`oliver.mueller@mpi-dortmund.mpg.de
`
`address:
`
`1. Introduction
`
`The diagnosis and the therapy of early-stage
`tumours have the potential
`to decrease morbidity
`and mortality of colorectal cancer patients. Invasive
`methods like sigmoidoscopy, colonoscopy and
`
`0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved.
`doi:10.1016/j.canlet.2004.12.011
`
`Geneoscopy Exhibit 1012, Page 1
`
`

`

`34
`
`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`roentgenography are valuable tools for the specific
`and sensitive detection and the exact localization of a
`colorectal tumour (reviewed in [1]). These techniques
`are unsuitable for broad screening programs mainly
`because of their low general acceptance [2]. The most
`common screening method is the faecal occult blood
`test (FOBT). Randomised trials revealed the clinical
`effectiveness in reduction of mortality through the
`implementation of the FOBT [3–5].
`Sidransky et al. introduced the analysis of faecal
`DNA for tumour specific alterations as a novel non-
`invasive diagnostic strategy [6]. Until now several
`studies have proven the usefulness and the feasibility
`of faecal DNA analysis (reviewed in [7]). Recent
`multi-target studies achieved sensitivities between 62
`and 91% and specificities up to 98%, respectively
`[8–10]. A few studies compared the outcomes of the
`molecular diagnosis (MD), which is the detection of
`tumour relevant alterations in faecal DNA, with these
`of
`the FOBT. The finding that 9 out of 11
`asymptomatic adenoma patients were identified by
`the analysis of five different DNA markers, whereas
`none was detected by FOBT,
`indicated a higher
`diagnostic sensitivity of the MD [9]. The data of a
`large prospective study, which is close to completion,
`will allow assessing the two methods and their
`combined application (annual meetings of
`the
`American College of Gastroenterology (ACG), Balti-
`more 2003 and of the American Gastroenterology
`Association (AGA), New Orleans 2004). We intended
`to judge in this retrospective study whether the
`analysis of three faecal DNA markers has the potential
`to complement or even to replace the FOBT and
`whether the combined application of the two methods
`might increase the overall diagnostic reliability.
`
`2. Materials and methods
`
`2.1. Study population and sample collection
`
`All patients and control donors provided their
`informed oral and written consent to provide samples.
`This study was approved by the responsible ethical
`committees at the Klinik fu¨r Chirurgie Marienhospital
`(Herne) and at the Berufsgenossenschaftliche Klini-
`ken Bergmannsheil (Bochum). None of the patients
`and none of
`the control donors had familial
`
`adenomatous polyposis coli or hereditary non-poly-
`posis colorectal cancer. Patients were diagnosed with
`colorectal tumours at the Berufsgenossenschaftliche
`Kliniken Bergmannsheil, the Marienhospital Herne or
`the Katholisches Krankenhaus Dortmund-West
`between 1998 and 2003. All tumours were diagnosed
`and located by imaging methods like colonoscopy and
`roentgenography. At least seven days after colono-
`scopy and between 2 and 4 days before surgical
`removal of the tumour 2 samples were taken from
`regular stools, immediately frozen at K20 8C and
`transferred to K80 8C within 24 h. The surgically
`removed tumours were classified according to the
`AJCC/TNM system (American Joint Committee on
`Cancer (AJCC) stages I through IV; Table 1) [11].
`Because of personnel alternations in the clinical
`departments in the course of the study, we could not
`reproduce the staging information for 12 tumours, the
`age and the gender of 8 patients and the location for 3
`tumours at the time of the analysis of the faecal
`samples. Control stool samples were from 24 female
`and 20 male donors, who volunteered to participate.
`The age group of control donors was 24–78 years
`(mean age 49). Control donors did not show any
`symptoms or signs of colorectal diseases as deter-
`mined in personal interviews.
`
`2.2. FOBT
`
`Two stool samples from each cancer patient and
`one sample from each control donor were analysed for
`the presence of occult blood by a standard guaiac test
`using commercial reagents and standard protocols
`(Care Diagnostica). Each FOBT was performed twice.
`A blue colour reaction within 120 s after the addition
`of the hydrogen peroxide solution was rated as a
`positive result. In case the double FOBT analysis of
`one stool sample or of two stool samples from one
`patient gave discrepant results, the sample was rated
`positive.
`
`2.3. MD
`
`Faecal DNA from all samples was purified using a
`method, which was optimised for faecal specimen
`[12]. An essential step of this method is the matrix-
`based elimination of PCR inhibiting and DNA
`damaging compounds (Stool Mini Kit, Qiagen).
`
`Geneoscopy Exhibit 1012, Page 2
`
`

`

`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`35
`
`Table 1
`Patient and tumor data of all samples and results obtained by the analysis of stool samples from cancer patients by FOBT and by MD
`
`Patient
`
`Gender/age
`
`AJCC/TNM
`
`FOBT
`
`MD positive markers
`
`79
`80
`83
`84
`87
`88
`89
`90
`92
`93
`95
`96
`121
`136
`138
`139
`140
`141
`172
`173
`174
`175
`176
`179
`180
`181
`182
`183
`184
`185
`186
`187
`188
`189
`190
`191
`192
`193
`251
`252
`253
`254
`255
`257
`274
`275
`276
`278
`283
`284
`285
`286
`
`M/70
`M/58
`M/67
`M/85
`n.a.
`F/76
`n.a.
`F/62
`F/53
`n.a.
`F/68
`n.a.
`M/51
`M/61
`M/65
`M/74
`F/73
`n.a.
`F/75
`M/73
`F/79
`F/75
`M/63
`M/68
`F/65
`F/92
`M/74
`M/60
`M/65
`M/75
`M/47
`M/73
`M/65
`M/76
`M/69
`M/76
`M/89
`F/64
`M/55
`M/74
`F/83
`M/66
`M/62
`n.a.
`F/70
`n.a.
`n.a.
`M/86
`M/82
`F/72
`M/81
`M/64
`
`n.a.*
`II/T3N0MxG2
`II/T3N0MxG2
`II/T3N0MxG2
`n.a.
`II/T3N0MxG2
`n.a.
`I/T2N0MxG2
`n.a.
`n.a.
`I/T2N0MxG2
`n.a.
`I/T2N0MxG2
`I/T2N0MxG2
`I/T2N0MxG2
`II/T3N0MxG2
`II/T3N0MxG2
`n.a.
`II/T3N0MxG2
`II/T3N0MxG2
`IV/T3N2M1hepG2
`II/T3N0MxG3
`IV/T3N1M1hepG3
`III/T3N1MxG2
`II/T4N0MxG2
`III/T4N1M0G2
`I/T2N0MxG2
`II/T3N0MxG2
`IV/T3N2M1hepG2
`I/T2N0MxG2
`III/T3N1MxG2
`II/T3N0MxG2
`IV/T3N1M1hepG2
`III/T3N2MxG2
`I/T2N0MxG2
`III/T3N1MxG2
`II/T3N0MxG2
`I/T2N0MxG2
`IV/T3N2M1G2
`II/T3N0MxG2
`III/T3N1MxG2
`II/T4N0MxG2
`III/T3N2MxG2
`n.a.
`I/T2N0MxG3
`n.a.
`n.a.
`n.a.
`III/T3N1MxG2
`n.a.
`II/T3NxMxG2
`IV/T3NxM1hepG2
`
`neg
`pos
`neg
`neg
`neg
`pos
`neg
`neg
`pos
`pos
`pos
`neg
`neg
`neg
`pos
`neg
`pos
`pos
`neg
`neg
`pos
`pos
`neg
`neg
`neg
`pos
`neg
`neg
`neg
`pos
`neg
`pos
`neg
`neg
`pos
`pos
`pos
`neg
`pos
`pos
`neg
`pos
`pos
`neg
`pos
`pos
`neg
`neg
`pos
`neg
`neg
`pos
`
`neg
`APC
`APC, L-DNA
`BAT26
`APC
`neg
`L-DNA
`L-DNA
`neg
`APC
`APC
`APC, L-DNA
`L-DNA
`APC
`neg
`neg
`neg
`APC
`APC
`APC
`APC, L-DNA
`neg
`APC
`APC
`APC
`neg
`APC
`L-DNA
`L-DNA
`BAT26
`L-DNA
`neg
`neg
`APC, L-DNA
`BAT26, L-DNA
`L-DNA
`L-DNA
`BAT26, L-DNA
`L-DNA
`neg
`L-DNA
`BAT26, L-DNA
`neg
`APC
`neg
`APC, L-DNA
`L-DNA
`L-DNA
`APC, L-DNA
`L-DNA
`neg
`neg
`
`(continued on next page)
`
`Geneoscopy Exhibit 1012, Page 3
`
`

`

`36
`
`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`Table 1 (continued)
`
`Patient
`
`Gender/age
`
`AJCC/TNM
`
`FOBT
`
`MD positive markers
`
`287
`289
`290
`292
`294
`
`M/92
`M/71
`M/66
`M/80
`F/83
`
`III/T3N1MxG3
`II/T4NxMxG3
`IV/T4N0M1
`I/T1N0MxG2R0
`III/T3N2MxG2
`
`pos
`pos
`pos
`pos
`pos
`
`neg
`neg
`neg
`neg
`neg
`
`n.a., data not available
`
`The quality of the purified DNA was validated by
`agarose gel electrophoresis. One tenth of the DNA
`obtained from one stool sample was used in a
`capturing step to enrich the human DNA from the
`DNA of non-human origin [13]. The overall yield of
`the capturing could be increased by the restriction
`digestion of the DNA with the enzyme DraI. To fish
`the APC gene out of the pool of genomic DNA,
`0
`10 pmol biotin labelled primer APC1 (5
`-TTAAAA-
`TATGCCACAGATATTCCT TCA TCACAGAAA-
`0
`) was added to 100 ml DraI digested stool
`CAGT-3
`DNA. After denaturing and re-annealing 10 ml
`streptavidin coupled magnetic beads (Dynal) were
`added and the mixture was incubated over night at
`room temperature. After washing, the beads coupled
`DNA was completely used as template in a first PCR.
`The APC gene region from codon 1069 through codon
`1984 including the mutation cluster region was
`analysed by PCR and by PTT of two overlapping
`fragments. The detailed experimental conditions of
`the following PTT have been published [13] (Hauss
`and Mu¨ller, The protein truncation test in mutation
`detection and molecular diagnosis, in: Methods in
`Molecular Biology, edited by Guido Grandi, The
`Humana Press,
`in press). Briefly,
`the primers of
`the first PCR coded the HA-tag, and the primers of the
`second PCR coded the Kozak motif and the T7
`promoter sequence, which are necessary for the
`following PTT. The first PCR was carried out at the
`following conditions: 95 8C (5 min)//[94 8C (30 s)/
`56 8C (30 s)/72 8C (90 s)] 20!//[94 8C (30 s)/58 8C
`(30 s)/72 8C (90 s)] 20!//72 8C (10 min) and under
`0
`the use of the following primers: APC fragment I: 5
`-
`CGCCATGTACCCCTACGACGTGCCCGACTAC
`0
`GCCTTCCAACCACATTTTGGACAGCAG-3
`and
`0
`5
`-CCGTCATTTTTCTGCCTCTTTCTCTTGGTT-
`0
`0
`3
`; APC fragment II: 5
`-CGCCATGTACCCCTAC-
`GACGTGCCCGACTACGCCGATGTGGAATTAA-
`0
`0
`GAATAATGCCT-3
`and 5
`-CCGTCATTTTCTT-
`
`0
`
`. One fifth of the
`TATTGTTGTTTTCTTGGTC-3
`volume of the first PCR was used in a second PCR at
`the conditions: 95 8C (5 min)//[94 8C (30 s)/60 8C
`(30 s)/72 8C (90 s)] 40!//72 8C (10 min) and using
`the following primers, which anneal within the
`0
`products of the first PCR: APC fragment I: 5
`-
`ATCCTAATACGACTCACTATAGGGAGCCAC-
`0
`0
`CATGTACCCCTACGACGTG-3
`and 5
`-CCGTCA-
`0
`TTTTTCTGCCTCTTTCTCTTGGTT-3
`; APC frag-
`0
`-ATCCTAATACGACTCACTATAGGG-
`ment II: 5
`0
`0
`AGCCACCATGTACCCCTACGACGTG-3
`and 5
`-
`CCGTCATTTTCTTTATTGTTGTTTTCTTGGTC-
`0
`3
`. Three of the control samples gave no reproducible
`results in the PCR reactions. All other faecal DNA
`samples from control donors and all samples from
`cancer patients gave robust PCR products. DNA
`purified from two stool samples from each cancer
`patient and from one sample from each control donor
`was analysed for the presence of alterations in 3
`different markers. Each positive result was confirmed
`in a repeated analysis. All samples showing signals,
`which were different to the wild-type controls, were
`further examined. For this the PCR fragments were
`cloned in a T/A cloning vector and the cloned gene
`fragments were sequenced by conventional methods.
`The analyses of all positive samples were repeated by
`starting with the purified DNA in order to limit the
`number of false positive results. The analysis of
`alterations in the microsatellite marker BAT26 and
`the detection of the relative amount of non-apoptotic
`L-DNA were performed according to published
`protocols [9,14]. All results are based on the analysis
`of at least two independently amplified PCR products.
`The primer sequences for L-DNA analysis were
`kindly provided by A.P. Shuber and D.A. Ahlquist.
`In case of discrepant MD results between the two
`different samples from one patient, DNA was
`extracted from a new faecal aliquot of the same
`patient and the analysis was repeated.
`
`Geneoscopy Exhibit 1012, Page 4
`
`

`

`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`37
`
`Table 2
`Localization of tumours relative to the splenic flexure of the colon
`and number of the corresponding stool samples diagnosed as
`positive by FOBT or by MD
`
`Site
`
`Proximal
`Distal
`n.a.
`Total
`
`N
`
`27
`27
`3
`57
`
`FOBT
`
`MD
`
`Combined use
`
`16
`11
`2
`29
`
`14
`21
`2
`37
`
`25
`26
`2
`53
`
`N, number of samples; n.a., at the time of analysis the information of
`the tumour localization was not available.
`
`2.4. Statistical evaluation
`
`Ninety-five percent confidence intervals were
`determined based on the exact binomial distribution.
`The Fisher exact test was used for comparison of the
`test results between patients with different tumour
`stages. To compare the single application of FOBT or
`MD with the combined application of both methods
`the McNemar’s test was used.
`
`3. Results
`
`The aim of this study was to determine the
`sensitivity and the specificity of the two non-invasive
`diagnostic methods, FOBT and MD, which is
`
`the analysis of faecal DNA for alterations in the
`three different markers APC, BAT26 and L-DNA. We
`analysed stool samples from 57 colorectal cancer
`patients and from 44 control donors. Twenty-seven
`tumours were located proximal and 27 tumours were
`located distal
`to the splenic flexure of the colon
`(Table 2). Twenty-eight patients had tumours, which
`did not spread to proximal or distant lymph nodes
`(stage I or II) at
`the time of staging diagnosis.
`Lymphatic spread was identified in 17 patients (stage
`III or IV) (Tables 1 and 3).
`Twenty-nine stool samples (51%; 95% confi-
`dence interval (CI): 38–63%) showed a positive
`FOBT test (FOBT pos). Thirty-seven faecal DNA
`samples (65%; CI: 52–76%) were MD positive
`(MD pos) demonstrating an alteration in at
`least
`one of the three tested DNA markers (Tables 1 and
`3). The MD positives include nineteen samples
`(33%) with truncated APC protein fragments in the
`PTT analysis (Fig. 1). We sequenced these positive
`samples and found mutations, which result in stop
`codons between codons 1254 and 1472 (Table 4).
`The sizes of the predicted translation products of
`PCR fragments carrying these stop codons correlate
`with the approximate sizes of
`the translation
`products, which were detected in the PTT
`(Fig. 1). The number of APC mutations detected
`in the samples was in the range of the numbers
`
`Table 3
`Summarized results of the analysis of the stool samples from cancer patients and sensitivity of MD and FOBT
`
`AJCC stage
`
`n.a.
`
`I
`
`II
`
`III
`
`IV
`
`Total
`
`N
`Mean age (range)
`Gender (male/female)
`n.a.
`FOBT pos
`MD pos
`APC
`BAT26
`L-DNA
`
`12
`70.3 (53–86)
`2/2
`8
`4
`10
`6
`0
`6
`
`11
`67.2 (51–80)
`7/4
`0
`6
`8
`3
`3
`4
`
`17
`72.7 (47–92)
`12/5
`0
`9
`9
`5
`2
`4
`
`10
`76.1 (47–92)
`7/3
`0
`6
`6
`3
`0
`5
`
`7
`65.3 (55–79)
`6/1
`0
`4
`4
`2
`0
`3
`
`MD pos or FOBT pos
`MD(APC or L-DNA) pos
`or FOBT pos
`MD pos and FOBT neg
`FOBT pos and MD neg
`
`11
`11
`
`7
`1
`
`11
`11
`
`5
`3
`
`15
`14
`
`6
`6
`
`10
`10
`
`4
`4
`
`6
`6
`
`2
`2
`
`57
`70.9 (47–92)
`34/15
`8
`29 (51%; 38–63%)
`37 (65%; 52–76%)
`19 (33%)
`5 (9%)
`22 (39%)
`
`53 (93%; 83–97%)
`52 (91%; 81–96%)
`
`24 (42%; 30–55%)
`16 (28%; 18–41%)
`
`The results, the relative sensitivities and the corresponding 95% confidence intervals are given for the FOBT, the MD and the combined
`application of the two methods. A sample was rated MD positive (MD pos), when one molecular marker was tested positive in the PTT, in the
`BAT26 or in the L-DNA analysis. n.a., AJCC staging of the corresponding tumours or age and gender were not available at the time of analysis.
`
`Geneoscopy Exhibit 1012, Page 5
`
`

`

`38
`
`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`Fig. 1. Positive PTT results of the APC gene analysis. Arrowheads point to bands, which represent the wild-type full-length protein (WT) or
`smaller proteins, which are truncated due to mutations. On the right side the sizes of marker fragments are indicated in kilodaltons. C, non-
`mutated control.
`
`described in other studies [9,15]. Ahlquist et al.
`(2000) detected APC mutations in 24% faecal DNA
`samples of cancer patients. Traverso et al. (2002)
`found APC mutations in 63% DNA samples of
`patients with T3N0M0 tumours. We detected
`BAT26 patterns different to the wild-type control
`in 5 (9%) samples (Fig. 2). Twenty-two (39%)
`samples showed high levels of L-DNA amplifica-
`tion products (Fig. 3). Thirty-five stool samples
`(61% of all samples or 95% of MD positive
`samples) showed alterations in the APC gene or in
`the relative L-DNA level. This high number reflects
`the high proportion of colorectal
`tumours, which
`harbour an APC gene mutation and a dysfunction
`in the regulation of apoptosis (Tables 1 and 3).
`Forty-two percent of the samples (CI: 30–55%)
`were positive in at least one of the MD markers, but
`gave negative FOBT results. Twenty-eight percent of
`the samples (CI: 18–41%) were positive exclusively
`in the FOBT, but not in the molecular test. When the
`results of both methods were combined, 53 (93%; CI:
`83–97%) samples were positive in FOBT or in
`molecular testing. The combined application of
`the two methods together leads to a significantly
`higher sensitivity than the application of any of the
`methods alone (P!0.01). When FOBT was combined
`with a limited MD, which is the analysis of the two
`molecular markers APC and L-DNA, a detection rate
`of 91% (CI: 81–96%) was achieved. The sensitivity of
`this combined application is higher than the sensi-
`tivity of the analysis by limited MD or by FOBT alone
`(P!0.01).
`
`In one control sample we found occult blood by
`FOBT resulting in an overall FOBT specificity of 98%
`(CI: 88–100%) (Table 5). In another control sample
`we detected a truncated APC protein fragment by
`PTT. The following sequence analysis revealed a C to
`T transition at nucleotide position 4117 leading to
`
`Table 4
`Sequence alterations in APC detected by sequencing of MD positive
`samples
`
`Sample
`
`Mutation at sequence position
`
`Resulting stop
`at codon
`
`80
`83
`87
`93
`95
`96
`136
`141
`172
`173
`174
`176
`179
`180
`182
`189
`257
`275
`283
`
`C4278del
`C4117T
`G4153T
`G4075T
`C4366T
`C4366T
`G3943T
`A3835T
`G3943T
`C3777del
`4409–4412del
`A4330del
`G4152^G4153insA
`C4117T
`C4117T
`T4310^G4311insGCCACCAA
`C4117T
`A4382del
`C4055A
`
`1472
`1367
`1379
`1353
`1450
`1450
`1309
`1273
`1309
`1254
`1472
`1472
`1414
`1367
`1367
`1472
`1367
`1472
`1346
`
`Nucleotide changes, the nucleotide sequence positions and the
`resulting stop codons are listed for the detected APC mutations. All
`faecal DNA samples were screened by PTT for alterations in APC.
`Positive samples were sequenced.
`
`Geneoscopy Exhibit 1012, Page 6
`
`

`

`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`39
`
`Fig. 2. Polyacrylamide gels for the analysis of alterations in the BAT26 microsatellite patterns. All BAT26 positive samples and 9 representing
`negative samples are shown. The upper (U) region of the gel contains PCR products representing the wild-type full-length products. Bands in
`the lower (L) regions of the gels indicate deletions in the BAT26 microsatellite. N, negative control without any DNA template; WT, wild-type
`control from normal tissue; P, positive control from tumour tissue containing the mutant BAT26.
`
`a stop signal in the corresponding codon 1367 (not
`shown). Both control donors showed no signs of
`malignancies or polyps according to colonoscopy
`analysis. Three samples from the control group did
`not give reproducible PCR results. In the final
`statistical calculation of the overall specificity these
`control samples were rated positive (Table 5). Thus
`the specificity of MD was 91% (CI: 79–96%), and the
`overall specificity of the combined application of
`FOBT and MD was 89% (CI: 76–95%).
`
`4. Discussion
`
`Tumour specific mutations in faecal DNA are
`supposed to be sensitive and specific markers for the
`non-invasive diagnosis of colorectal cancer (reviewed
`in [7,16]).
`In this study the sensitivity of
`the
`retrospective diagnosis of colorectal carcinoma by
`the analysis of the three different molecular markers
`APC, BAT26 and L-DNA was 65%. The sensitivities
`described by others for the analysis of five different
`markers reach from 64 to 90% [9,10]. The differences
`
`between the sensitivities might be due to variations in
`sample quality, in the applied analytical methods and
`in the laboratory-specific handling of the samples. The
`high potential of the MD became obvious, when we
`compared the MD outcomes with those of the most
`common non-invasive screening test, the FOBT. The
`MD showed a 14% higher detection rate than the
`FOBT, and 42% of all samples were negative in
`the FOBT, but still detectable by MD. Nevertheless,
`our data also confirm the value of the FOBT.
`Transferring the data to a putative random non-
`invasive screen 28 out of 100 cancer patients would be
`detected only by FOBT but not by MD. The
`sensitivity of MD can be even further increased up
`to 93% by the additional application of FOBT. While
`the combination of the FOBT and MD leads to a
`significant increase in sensitivity, it leads only to a
`marginal decrease in specificity (Table 5).
`There was no significant correlation of the FOBT
`or MD results with the gender, the age or the tumour
`stages. This finding is in line with the results of others
`[9]. Furthermore, we did not find a significant
`correlation of the positive markers with the location
`
`Fig. 3. Agarose gels for the analysis of the amplification of L-DNA. The two gels show the analysis of PCR of KRAS-L-DNA from 9
`representing stool DNA samples, which were amplified in duplicates (a, b). M, Marker; N, negative control without any DNA; P, positive
`control.
`
`Geneoscopy Exhibit 1012, Page 7
`
`

`

`40
`
`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`Table 5
`The numbers of positive samples and Summary of results
`
`Cancer patients
`
`Control donors
`
`Sensitivity (%)
`
`Specificity (%)
`
`FOBT pos
`MD pos
`MD pos or FOBT pos
`MD(APC or L-DNA)
`pos or FOBT pos
`
`29
`37
`53
`52
`
`1
`4
`5
`5
`
`51 (38–63)
`65 (52–76)
`93 (83–97)
`91 (81–96)
`
`98 (88–100)
`91 (79–96)
`89 (76–95)
`89 (76–95)
`
`The relative sensitivities, specificities and the corresponding 95% confidence intervals of the MD, the FOBT and of the combined application of
`the two methods are given. The MD positive samples of the control donors include three samples, which could not be analysed.
`
`of the tumours (Table 2). The findings of others, who
`described indications for a correlation between
`tumour location and BAT26 and L-DNA markers,
`have to be confirmed by the analysis of a high number
`of samples [9].
`In order to determine the specificity we analysed
`the samples of randomly chosen control donors, who
`showed no obvious symptoms of colorectal diseases.
`This strategy might have led to an overestimated
`specificity in comparison to an authentic screening,
`which includes both non-symptomatic and sympto-
`matic cancer patients and healthy donors. Because the
`analysed MD markers APC and BAT26 correlate
`directly to tumour development and do not necessarily
`correlate with the onset of symptoms, the analysis of
`these two markers should not distort
`the overall
`specificity. We are still aware that
`the FOBT
`outcomes might have led to the overestimation of
`the overall specificity. A subsequent study, which will
`include samples from non-symptomatic as well as
`samples from symptomatic individuals, will clarify
`this aspect.
`This is one of the first studies, which evaluate and
`directly compare the potential of the two non-invasive
`methods MD and FOBT in the diagnosis of colorectal
`carcinomas. The combined application of the two
`methods together
`led to a significantly higher
`sensitivity than the application of any of the methods
`alone. Regarding the low cost and the little technical
`expense of the FOBT, we suggest that MD and FOBT
`should complement rather than replace each other.
`The sensitivity of the combined application of MD
`and FOBT is in the range reported for invasive
`methods. This leads to the question whether both
`methods should be used together in broad diagnostic
`screening programs. We are aware that our
`
`retrospective data do not allow any conclusion on
`the potential of this approach in diagnostic screening.
`Prospective comparative studies, which include a very
`high number of individuals, have to prove the value of
`the combined application of MD and FOBT. Prelimi-
`nary results of a large study were presented in abstract
`form (ACG and AGA annual meetings). The final
`results of this study will allow further conclusions
`about the effectiveness of the two methods and their
`combined application.
`
`Acknowledgements
`
`We thank Alfred Wittinghofer for generous sup-
`port. We acknowledge the primer sequences for the
`L-DNA analysis provided by A.P. Shuber and
`D.A. Ahlquist.
`
`References
`
`[1] C.J. Kahi, D.K. Rex, Current and future trends in colorectal
`cancer screening, Cancer Metastasis Rev. 23 (2004) 137–144.
`[2] A.L. Frazier, G.A. Colditz, C.S. Fuchs, K.M. Kuntz, Cost-
`effectiveness of screening for colorectal cancer in the general
`population, J. Am. Med Assoc. 284 (2000) 1954–1961.
`[3] J.S. Mandel,
`J.H. Bond, T.R. Church, D.C. Snover,
`G.M. Bradley, L.M. Schuman, F. Ederer, Reducing mortality
`from colorectal cancer by screening for fecal occult blood.
`Minnesota Colon Cancer Control Study, N. Engl. J. Med. 328
`(1993) 1365–1371.
`J.O. Chamberlain, M.H. Robinson,
`[4] J.D. Hardcastle,
`S.M. Moss, S.S. Amar, T.W. Balfour, et al., Randomised
`controlled trial of faecal-occult-blood screening for colorectal
`cancer, Lancet 348 (1996) 1472–1477.
`
`Geneoscopy Exhibit 1012, Page 8
`
`

`

`N. Kutzner et al. / Cancer Letters 229 (2005) 33–41
`
`41
`
`Jorgensen,
`J. Olsen, O.D.
`[5] O. Kronborg, C. Fenger,
`O. Sondergaard, Randomised study of screening for colorectal
`cancer with faecal-occult- blood test, Lancet 348 (1996) 1467–
`1471.
`[6] D. Sidransky, T. Tokino, S.R. Hamilton, K.W. Kinzler,
`B. Levin, P. Frost, B. Vogelstein,
`Identification of
`ras
`oncogene mutations in the stool of patients with curable
`colorectal tumors, Science 256 (1992) 102–105.
`[7] K.S. Tagore, T.R. Levin, M.J. Lawson, The evolution to stool
`DNA testing for colorectal cancer, Aliment. Pharmacol. Ther.
`19 (2004) 1225–1233.
`[8] D. Calistri, C. Rengucci, R. Bocchini, L. Saragoni, W. Zoli,
`D. Amadori, Fecal multiple molecular tests to detect colorectal
`cancer in stool, Clin. Gastroenterol. Hepatol. 1 (2003) 377–
`383.
`[9] D.A. Ahlquist, J.E. Skoletsky, K.A. Boynton, J.J. Harrington,
`D.W. Mahoney, W.E. Pierceall, et al., Colorectal cancer
`screening by detection of altered human DNA in stool:
`feasibility of a multitarget assay panel, Gastroenterology 119
`(2000) 1219–1227.
`[10] K.S. Tagore, M.J. Lawson, J.A. Yucaitis, R. Gage, T. Orr,
`A.P. Shuber, M.E. Ross, Sensitivity and specificity of a stool
`
`DNA multitarget assay panel for the detection of advanced
`colorectal neoplasia, Clin. Colorectal. Cancer 3 (2003) 47–
`53.
`[11] F.L. Greene, D.L. Page, I.D. Fleming, A. Fritz, C.M. Balch,
`D.G. Haller, M. Morrow,, sixth ed. AJCC Cancer Staging
`Manual, XIV, 2002.
`[12] R. Deuter, S. Pietsch, S. Hertel, O. Muller, A method for
`preparation of fecal DNA suitable for PCR, Nucleic Acids Res.
`23 (1995) 3800–3801.
`[13] S. Kahmann, P. Herter, C. Kuhnen, K.M. Muller, G. Muhr,
`D. Martin, et al., A non-radioactive protein truncation test for
`the sensitive detection of all stop and frameshift mutations,
`Hum. Mutat. 19 (2002) 165–172.
`[14] G. Traverso, A. Shuber, L. Olsson, B. Levin, C. Johnson,
`S.R. Hamilton, et al., Detection of proximal colorectal cancers
`through analysis of faecal DNA, Lancet 359 (2002) 403–404.
`[15] G. Traverso, A. Shuber, B. Levin, C. Johnson, L. Olsson,
`D.J. Schoetz, et al., Detection of APC mutations in fecal DNA
`from patients with colorectal tumors, N. Engl. J. Med. 346
`(2002) 311–320.
`[16] D.A. Ahlquist, A.P. Shuber, Stool screening for colorectal
`cancer: evolution from occult blood to molecular markers,
`Clin. Chim. Acta 315 (2002) 157–168.
`
`Geneoscopy Exhibit 1012, Page 9
`
`