Reles, Angela : MOLECULAR GENETIC ALTERATIONS IN OVARIAN CANCER The Role of the p53 Tumor Suppressor Gene and the mdm2 Oncogene

72

Kapitel 4. RESULTS

4.1 Alterations of the p53 tumor suppressor gene in ovarian cancer

4.1.1 p53 mutations

In a cohort of 178 ovarian carcinomas we found a total of 145 p53 sequence alterations by PCR-SSCP and direct sequencing (66%) (Table 7). Among the sequence alterations, 100 p53 mutations were identified in 99 cases, one of which had a mutation in both exon 7 and exon 8. In addition, four silent mutations without alteration of the amino acid sequence were identified and were considered polymorphisms. Forty-five of the sequence alterations were polymorphisms or intron alterations of undetermined significance.

Table 7: Summary of p53 sequence alterations in 178 ovarian carcinomas

Type of sequence alteration

n

% cases

 

 

 

 

 

 

Mutations with amino acid alterations in exon 2-11

92

51.1

(in 91 cases, double mutation in case #2341)

 

 

 

 

 

Splice site mutations

8

4.5

 

 

 

Polymorphisms in exon 2-11

 

 

exon 4, codon 36

1

0.6

exon 4, codon 72

9

5.1

exon 6, codon 213

1

0.6

exon 6, codon 224

1

0.6

exon 7, codon 231

1

0.6

 

 

 

Intron alterations (excluding splice site mutations)

 

 

intron 3 polymorphism (16 bp repeat)

24

13.5

intron 6 alteration (nt 13964)

2

1.1

intron 10 alteration (nt 17708)

2

1.1

intron 10 alteration (nt 18550)

4

2.3

Total number of sequence alterations (in 117 cases)

145

65.7


73

Fig. 10 A-C: p53 mutation and overexpression
A:
Single Strand Conformation Polymorphism (SSCP) analysis of the p53 gene, exon 5 in ovarian carcinomas.
Arrows in lanes 7 and 8 indicate band shifts suspicious for mutations.
B: DNA sequence analysis of exon 5 of the p53 gene in endometrioid ovarian carcinoma #2335.
Missense mutation in codon 141, TGC to TGG (amino acid exchange cystein to tryptophan).
C: Immunostaining shows overexpression of the p53 protein in the tumor nuclei (case #2335).


74

Fig. 11 A-C: p53 mutation and overexpressio
A:
Single Strand Conformation Polymorphism (SSCP) analysis of p53, exon 9 in ovarian carcinomas. Arrows in lanes 3 and 10 (case #3402) indicate band shifts suspicious for mutations.
B: DNA sequence analysis of exon 9 of the p53 gene in endometrioid ovarian carcinoma #3402. A deletion mutation of one basepair in codon 320 causes a frameshift.
C: Immunostaining shows overexpression of the p53 protein in the tumor nuclei (case #2335.)


75

Thirteen of the p53 sequence alterations considered polymorphisms were exon polymorphisms and 32 were intron alterations, some of which are known as polymorphisms. Nine sequence alterations, which are known as a Codon 72 ArgrarrPro polymorphism, were found in exon 4.

One hundred p53 mutations were identified by PCR-SSCP and direct sequencing in 99 ovarian cancer patients (56%) (Fig. 10-12, Table 8). 64 of these alterations have been previously described in a study of 105 ovarian carcinoma patients (Wen et al. 1999). One ovarian cancer patient had a mutation in exon 7 and exon 8, both with alteration of the amino acid sequence. The majority of mutations (86%) was found in exon 5 (22 mutations), exon 6 (12 mutations), exon 7 (26 mutations) and exon 8 (28 mutations) (Fig. 11, Table 8). However, 14% of the mutations were found outside codons 126-306 in exon 4 (3 mutations), exon 9 (3 mutations), intron 4 (3 splice junction mutations), intron 5 (1 splice junction mutation), intron 6 (1 splice junction mutation), intron 7 (2 splice site muations), and intron 8 (1 splice junction mutations).

Fig. 12: p53 mutations in ovarian cancer. Mutations of the p53 gene were identified in 99/178 epithelial ovarian carcinomas. The majority of mutations were missense mutations.

Out of 100 mutations, 72% were missense mutations (Fig. 12), 7% were single base substitutions which resulted in the introduction of a premature stop codon, 3% were insertions of one to six basepairs, 10% were deletions of one to seven basepairs and 8% were splice junction mutations (Table 8). Four „silent„ DNA sequence alterations, which did not result in an amino acid alteration, were considered polymorphisms and were located in exon 4 (codon 36), exon 6 (codon 213 and 224), and exon 7 (codon 231).


76-77

Table 8: p53 mutations and protein overexpression in ovarian carcinomas

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Case

Histo-

SSCP

Exon

Codon

Amino

DNA

Amino Acid

Mutation

Im**

TS/TV***

#

patho-

Result

 

#

Acids

Sequence

Sequence

Type

p53

CpG

 

logy

 

 

 

Cons*

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3379

Ser

+

4

36

NC

CCG to CCA

Pro / Pro

Polymorph

+

(TS)

2722

Ser

+

4

42

NC

GAT to TAT

Asp to Tyr

Missense

+

TV

3401

Ser

+

4

49, 50

NC

GAT ATT to GA..TT

frameshift

2 bp Del

-

-

715

Ser

-

4

73

NC

GTG to TTG

Val to Leu

Missense

+

TV

3391

Endo

+

4

nt 12301

Co

CGgt to CGga

intron 4

5' splice jun

-

TV

 

 

 

 

 

 

 

 

 

 

 

2658

Undiff

+

5

nt 13053

Co

agTA to ggTA

intron 4

3' splice jun

+

TS

3355

Endo

+

5

nt 13053

Co

agTA to ggTA

intron 4

3' splice jun

+

TS

1557

Ser

+

5

130

Co

CTC to GTC

Leu to Val

Missense

+

TV

3400

Ser

+

5

132

Co

AAG to AGG

Lys to Arg

Missense

+

TS

1282

Ser

+

5

132

Co

AAG to ACG

Lys to Thr

Missense

+

TV

ov 70

Ser

+

5

135

Co

TGC to TGG

Cys to Trp

Missense

+

TV

3367

Ser

+

5

136

Co

CAA to TAA

Gln to Stop

Nonsense

-

TS

687

Ser

+

5

136

Co

CAA to TAA

Gln to Stop

Nonsense

-

TS

2335

Endo

+

5

141

Co

TGC to TGG

Cys to Trp

Missense

+

TV

2745

Ser

+

5

151

NC

CCC to CAC

Pro to His

Missense

+

TV

3389

Mixed

+

5

158

NC

CGC to CCC

Arg to Pro

Missense

+

TV

824

Mixed

+

5

159

NC

GCC to GTC

Ala to Val

Missense

+

TS;CpG

704

Ser

+

5

160

NC

ATG to AAG

Met to Lys

Missense

+

TV

769

Ser

+

5

163

NC

TAC to TGC

Tyr to Cys

Missense

+

TS

2354

Brenner

+

5

163

NC

TAC to TGC

Tyr to Cys

Missense

+

TS

713

Ser

+

5

166

NC

TCA to TGA

Ser to Stop

Nonsense

-

TV

2650

Undiff

+

5

175

Co

CGC to CAC

Arg to His

Missense

+

TS;CpG

3371

Mixed

+

5

175

Co

CGC to CAC

Arg to His

Missense

+

TS;CpG

762

Ser

+

5

175

Co

CGC to CAC

Arg to His

Missense

+

TS;CpG

1270

Ser

+

5

175

Co

CGC to CAC

Arg to His

Missense

+

TS;CpG

3354

Ser

+

5

179

Co

CAT to GAT

His to Asp

Missense

+

TV

3529

Ser

+

5

179

Co

CAT to TAT

His to Tyr

Missense

+

TS

698

Adenoca

+

5

179

Co

CAT to CGT

His to Arg

Missense

+

TS

2716

Mixed

+

5

179

Co

CAT to CGT

His to Arg

Missense

+

TS

2721

Ser

+

5

nt 13240

-

TGgt to TGgc

intron 5

5'-splice jun

-

TS

 

 

 

 

 

 

 

 

 

 

 

3353

undiff

+

6

192

NC

CAG to TAG

Gln to Stop

Nonsense

-

TS

ov 59

Ser

+

6

193

NC

CAT to CGT

His to Arg

Missense

+

TS

2719

Ser

+

6

193-195

NC

CATCTTATC to GCCCCT

His,Leu,Ile to Ala, Pro

Del / Ins

+

-

2309

Undiff

+

6

195

NC

ATC to AAC

Ile to Asn

Missense

+

TV

805

Ser

+

6

201

NC

TTG to TTT

Leu to Phe

Missense

+

TV

2352

Ser

+

6

203

NC

GTG to TTG

Val to Leu

Missense

+

TV

2733

Ser

+

6

213

NC

CGA to CTA

Arg to Leu

Missense

-

TV

2340

Ser

+

6

213

NC

CGA to CGG

Arg / Arg

Polymorph

-

(TS)

2737

Ser

-

6

215

NC

AGT to GT

frameshift

1bp Del/Nonsense

-

-

2349

Ser

+

6

216

NC

GTG to TTG

Val to Leu

Missense

+

TV

241

Ser

-

6

220

NC

TAT to TGT

Tyr to Cys

Missense

+

TS

808

Clear

-

6

224

NC

GAG to GAA

Glu / Glu

Polymorph

-

(TS)

 

 

 

 

 

 

 

 

 

 

 

3373

Ser

+

7

nt 13999

-

agGT to aaGT

intron 6

3' splice jun

+

TS

ov 49

Ser

+

7

227

NC

TCT to TCA

Asp to Ser

Missense

+

TV

2705

Mixed

+

7

231

NC

ACC to ACA

Thr / Thr

Polymorph

+

(TV)

3379

Ser

+

7

234

Co

TAC to TCC

Tyr to Ser

Missense

+

TV

2344

Clear

+

7

234

Co

TAC to TGC

Tyr to Cys

Missense

-

TS

2717

Ser

+

7

234

Co

TAC to TGC

Tyr to Cys

Missense

+

TS

2656

Endo

+

7

237

Co

ATG to ATT

Met to Ile

Missense

+

TV

3344

Endo

+

7

239 / 240

Co

AAC to AAC TTA

Leu ins

3 bp Ins

+

-

2659

Endo

+

7

242

Co

TGC to TAC

Cys to Tyr

Missense

+

TS

790

Mixed

+

7

242

Co

TGC to TTC

Cys to Phe

Missense

+

TV

2338

Endo

+

7

242

Co

TGC to TTC

Cys to Phe

Missense

+

TV

2724

Endo

+

7

242

Co

TGC to AGC

Cys to Ser

Missense

+

TV

2730

Ser

+

7

243

Co

ATG to GTG

Met to Val

Missense

+

TS

665

Ser

+

7

244

Co

GGC to AGC

Gly to Ser

Missense

+

TS

1752

Muc

+

7

245

Co

GGC to AGC

Gly to Ser

Missense

+

TS;CpG

1559

Ser

+

7

245

Co

GGC to GAC

Gly to Asp

Missense

+

TS;CpG

3540

Ser

+

7

248

Co

CGG to CAG

Arg to Glu

Missense

+

TS;CpG

802

Endo

+

7

248

Co

CGG to TGG

Arg to Trp

Missense

+

TS;CpG

249

Ser

+

7

248

Co

CGG to CAG

Arg to Gln

Missense

+

TS;CpG

3403

Undiff

+

7

249, 250

Co

AGG CCC to AGT TCC

Arg,Pro to Ser,Ser

Dble-Missense

+

TV,TS

2364

Ser

+

7

251, 252

Co

ATC CTC to ATC

Leu del

3 bp Del

+

-

1589

Ser

+

7

253

Co

ACC to CCC

Thr to Pro

Missense

+

TV

2651

Undiff

+

7

254 - 256

Co

ATC ATC ACA to A

frameshift

7 bp Del

-

-

2001

Ser

+

7

255

Co

ATC to AGC

Ile to Ser

Missense

-

TV

2337

Endo

+

7

258

Co

GAA to TAA

Glu to Stop

Nonsense

+

TV

2341

Ser

+

7

258

Co

GAA to TAA

Glu to Stop

Nonsense

-

TV

3362

Undiff

+

7

259

NC

GAC to TAC

Asp to Tyr

Missense

+

TV

3347

Ser

+

7

nt 14110

-

AGgt to AGtt

intron 7

5' splice jun

-

TV

 

 

 

 

 

 

 

 

 

 

 

2725

Ser

+

8

nt 14449

-

tagTG to aagTG

intron 7

3' splice jun

+

TV

767

Ser

+

8

264

NC

CTA to TA

frameshift

1 bp Del

+

-

696

Ser

+

8

264, 265

NC

CTA CTG to CTG

Leu,Leu to Leu

3 bp Del

+

-

3364

Muc

+

8

266

NC

GGA to CGA

Gly to Arg

Missense

+

TV

ov45

Ser

+

8

266

NC

GGA to AGA

Gly to Arg

Missense

+

TS

2341

Ser

+

8

267

NC

CGG to CCG

Arg to Pro

Missense

-

TV

2332

Endo

+

8

267

NC

CGG to CCG

Arg to Pro

Missense

+

TV

2708

Mixed

-

8

271

Co

GAG to TAG

Glu to Stop

Nonsense

-

TV

2155

Ser

-

8

272

Co

GTG to ATG

Val to Met

Missense

+

TS

2732

Ser

-

8

272

Co

GTG to TTG

Val to Leu

Missense

+

TV

3346

Ser

+

8

273

Co

CGT to CAT

Arg to His

Missense

+

TS;CpG

3358

Undiff

+

8

273

Co

CGT to CAT

Arg to His

Missense

+

TS;CpG

3382

Endo

+

8

273

Co

CGT to CTT

Arg to Leu

Missense

+

TV

2715

Ser

-

8

273

Co

CGT to TGT

Arg to Cys

Missense

+

TS;CpG

693

Ser

+

8

273

Co

CGT to TGT

Arg to Cys

Missense

+

TS;CpG

700

Ser

-

8

273

Co

CGT to GGT

Arg to Gly

Missense

+

TV

2718

Ser

+

8

275

Co

TGT to TAT

Cys to Tyr

Missense

+

TS

ov 93

Ser

+

8

275

Co

TGT to GGT

Cys to Gly

Missense

+

TV

ov 164

Ser

+

8

275

Co

TGT to GGT

Cys to Gly

Missense

+

TV

2326

Mixed

+

8

276

Co

GCC to GTC

Ala to Val

Missense

+

TS

3385

Endo

+

8

280

Co

AGA to AGT

Arg to Ser

Missense

+

TV

2328

Ser

+

8

280

Co

AGA to ATA

Arg to Ile

Missense

+

TV

1751

Ser

+

8

281

Co

GAC to GAA

Asp to Glu

Missense

+

TV

2182

Endo

+

8

281

Co

GAC to GAG

Asp to Glu

Missense

-

TV

ov 72

Ser

+

8

281

Co

GAC to CAC

Asp to His

Missense

+

TV

3351

Ser

+

8

282

Co

CGG to TGG

Arg to Trp

Missense

+

TS;CpG

1756

Clear

+

8

282

Co

CGG to TGG

Arg to Trp

Missense

+

TS;CpG

1254

Ser

+

8

286

Co

GAA to GTA

Glu to Val

Missense

+

TV

3386

Ser

+

8

300

NC

CCC to ..CC

frameshift

1 bp Del

+

-

3461

Endo

+

8

301

NC

CCA to CCC A

frameshift

1 bp Ins

-

-

 

 

 

 

 

 

 

 

 

 

 

2339

Ser

+

9

nt 14680

-

agCA to aaCA

intron 8

3' splice jun

-

TS

1749

Undiff

+

9

 

NC

AAC to AA..

frameshift

1 bp Del

-

-

3402

Endo

+

9

 

NC

AAG to A..G

frameshift

1 bp Del

+

-

3394

Undiff

+

9

324

311

GAT to G..T

frameshift

1 bp Del

+

-

 

 

 

 

 

320

 

 

 

 

 

* according to Cho et al. 1994
** Im p53: Immunostaining for p53
*** TS / TV: Transition / Transversion
Histopathology: Ser: Serous, Endo: Endometrioid, Undiff: Undifferentiated
Muc: Mucinous, Clear: Clear Cell, Mixed: Mixed Epithelial


78

Among the cases with insertions, one had a nine basepair deletion and a six basepair insertion (case #2719). One case had a 3 bp insertion of TTA resulting in the insertion of the amino acid residue Leucin between codon 239 and 240 in exon 7 (#3344). A one basepair insertion (case #3461) at codon 301, exon 8 caused a frameshift with introduction of a premature stop at codon 305.

Eleven cases had deletions of 1-9 bp, four of which (cases # 2719, 2737, 767, 696) have been described previously (Wen et al. 1999). Of the newly identified deletions, four were 1 bp deletions, one in exon 8 (case # 3386) and three in exon 9 (case # 1749, 3402, 3394), all resulting in the introduction of a premature stop codon at codon 344. Three of these cases had p53 protein overexpression. One case had a 3 bp deletion in exon 7, resulting in the loss of a leucine residue (case #2364) and one case had a 2 bp deletion in exon 4, causing the introduction of a stop codon at codon 50 (case #3401). This case was negative in immunohistochemistry. A seven bp deletion in exon 7 (case #2651), affecting codons 254-256, resulted in a frameshift affecting the entire sequence between codon 256 and a premature stop at codon 344.

Furthermore, we identified 5 splice junction mutations (cases # 3391, 2658, 3355, 3373, 3347) in addition to three cases described in our previous study (Wen et al. 1999). Two of these affected the donor site at exon 4/intron 4 and exon 7/intron 7 respectively (Table 8). Both cases were negative for immunohistochemistry. Three splice site mutations involved the consensus acceptor site at intron 4/exon5 (2 cases) and intron 6/exon 7. These three cases had p53 protein overexpression (Table 8).

Four regions of the translated sequence of p53 have been highly conserved throughout evolution and span codon 117-124 (II), 171-181 (III), 234-258 (IV) and 270-286 (V). These regions span a total of 79 codons. In 62% (62/100) of the cases, mutations were located in these highly conserved regions (Fig. 13). Region II, which spans part of exon 4 and exon 5, contained 10 mutations including 3 splice site mutations in intron 4 (10%). Region III contained 8 mutations (8%), region IV 23 mutations (23%) and region V a total of 21 mutations (21%). Thirty-eight percent of the mutations were found in non-conserved regions of the gene. The codons most frequently mutated were codons 175, 179, 234, 242, 248, 273, 275 and 281 which were each mutated in three or more different carcinomas (Fig. 13).


79

Fig. 13: p53 mutations in evolutionary highly conserved domains of the gene. The upper part of the figure shows the number of mutations in codons 36 through 324 of the p53 gene. Mutations in evolutionary highly conserved regions are marked in yellow, mutations in non-conserved regions in dark blue and splice site mutations in white. The lower part of the figure shows the corresponding DNA interacting regions of the p53 protein.


80

Fourty-two mutations (42%) were classified as transition mutations, 46 (45%) as transversions, and 13% were deletions or insertions. We found 29 G:C to A:T transitions and 13 A:T to G:C transitions. Among the G:C to A:T transitions 16 (55%) were located in CpG sites, which are known to be potential sites of DNA methylation. Out of 46 transversions, we found 20 G:C to T:A mutations, 12 G:C to C:G, 8 T:A to A:T and 6 A:T to G:C mutations.

4.1.2 p53 polymorphisms and intron alterations

A total number of 45 p53 polymorphisms and intron alterations of unknown significance were found in 33 cases (19%). In 15/33 cases (46%) with polymorphisms or intron alterations additional p53 mutations with an amino acid exchange had been identified. This accounted for 15% of the cases with mutations. Only 61/178 ovarian cancer cases (34%) were found to have a completely normal p53 sequence with neither mutations nor other sequence alterations in the DNA segments which were analyzed.

The most frequent exon polymorphism was located in exon 4 at codon 72 (Fig. 14), and was found in 9 cases (5.1%). This CGCrarrCCC polymorphism at codon 72 results in an arginine to proline amino acid exchange (Harris et al. 1986, Matlashewski et al. 1987). In one case (#3379), an additional CCG to CCA (Pro/Pro) polymorphism was identified at codon 36 of exon 4 (Table 9 and 10).

A further exon polymorphisms, CGArarrCGG (Arg/Arg), was identified in exon 6 at codon 213. Another silent mutation which was considered a polymorphism, occured at codon 224 of exon 6 as a GAGrarrGAA (Glu/Glu) polymorphism without amino acid alteration. In exon 7 an ACCrarrACA (Thr/Thr) polymorphism was identified at codon 231 (Table 9 and 10).

An intron 3 polymorphism (Fig. 15), which is known as a 16 bp repeat (Lazar et al. 1993) at nucleotide 11951, was found in 24 (14%) of the cases. Six (25%) of the cases were identified by SSCP as homozygote and 18 (75%) were heterozygote for this intron 3 alteration. Five cases were sequenced to confirm the polymorphism suspected by the SSCP alteration.

Eight p53 sequence alterations were identified in intron sequences, flanking the exons 7, 10 and 11 (Table 9 and 10). In all of these cases, a shift in the SSCP bands was noticed without evidence of a mutation in the exon sequence. In intron 6 at position 13964, a grarrc base exchange was identified in two cases. No normal tissue was available in these cases to verify the polymorphism.


81

Fig. 14: CGC rarr CCC (Arg rarr Pro) polymorphism at codon 72 of the p53 gene
A: SSCP analysis of p53, exon 4 in ovarian carcinomas. DNA in lanes 2,4,5,6,and 8 shows
band shifts suspicious for sequence alterations and was analyzed by DNA sequencing.
B: DNA sequence analysis of p53, exon 4 shows a CGC to CCC poymorphism at codon 72,
resulting in an arginine to proline amino acid exchange.


82

Fig. 15: Intron 3 polymorphism of the p53 gene
A:
Single Strand Conformation Polymorphism (SSCP) of p53, exon 3 with the adjacent intron
sequence shows a band shift in lanes 2,4,8 and 9 indicating the intron 3 polymorphism.
B: Wild type (lanes 3 and 4) and the polymorphism sequence (lanes 1 and 2) of p53, intron 3
with a 16 bp insertion (lanes 1 and 2) and adjacent sequence are shown as
5‘ to 3‘: GCTggggacctggagggctGGGGGG


83

In intron 10 at position 17708, an ararrt base exchange was found in 2 out of five cases which had shown SSCP alterations in exon 10. In 4 out of 9 cases which had shown an obvious band shift in the SSCP analysis for exon 11, a crarrt base exchange at position 18550 was identified.

Multiple polymorphisms were identified in 11/178 (6%) of all ovarian cancer cases and 11/33 (33%) of the cases with polymorphisms. In case #3379, three polymorphisms were identified, two in exon 4 (codon 36 and codon 72) and one in intron 3. The other cases had two polymorphisms. Eight out of nine cases with exon 4 polymorphisms (89%) and 3/6 cases with intron 10 alterations (50%) had an additional intron 3 polymorphism.

4.1.3 Confirmation of p53 polymorphisms in normal tissue DNA

In cases with exon 4, exon 6 or exon 7 polymorphisms, and in cases with intron 6 or intron 10 alterations, normal tissue of the ovarian cancer patients, which had been formalin fixed and paraffin embedded at the time of surgery, was sequenced at the same DNA site (Table 10). Normal tissue was available in 12/20 of these cases. In three cases with a codon 72 ArgrarrPro polymorphism, the same sequence alteration was found in normal tissue. In two heterozygote cases, in which the CGC sequence was stronger than the CCC sequence and in one case with equally strong G and C band only, the CGC (arginin) sequence was identified in normal tissue. In two cases with a homozygote CGCrarrCCC polymorphism in the tumor, only CGC was identified in normal tissue. In the cases with exon 6, intron 6 and exon 7 polymorphisms no normal tissue was available

The intron 10 alteration at nucleotide 17708 was confirmed in one case, in which normal tissue was available. We have identified this alteration in two ovarian carcinoma cases, which had shown SSCP band shifts. We assume that this is an ararrt polymorphism at nucleotide 17708, which has not been described so far. Furthermore, the intron 10 crarrt sequence alteration at nucleotide 18550 was found in four ovarian carcinoma cases and could be confirmed in all three cases, in which normal tissue had been analyzed (Table 10). We therefore consider these intron alterations as two novel intron 10 polymorphisms.

4.1.4 p53 protein overexpression

p53 protein overexpression, using a cut-point of ge10% positively stained nuclei, was found in 62% (110/178) of the cases. The cut-off for p53 overexpression was chosen as 10%, because we found frozen sections from proliferating normal


84

Fig. 16: Intron 10 polymorphism of the p53 gene
at nucleotide 17708
A novel polymorphism of the p53 gene was identified
at nucleotide 17708 (a>t) of intron 10 by SSCP analysi
and DNA sequencing in the ovarian cancer case #3402.


85

Fig. 17: Intron 10 polymorphism of the p53 gene at nucleotide 18550
A:
SSCP analysis of p53 exon 11 and flanking intron sequence in ovarian carcinomas.
The band shift in lane 5 (case #3394) is suspicious for DNA sequence alterations.
B: DNA sequence analysis of p53 exon 11/ intron 10 reveals a novel c>t polymorphism
at nucleotide 18550 of intron 10 in the ovarian cancer cases #2660 and #3344.
The polymorphism was confirmed in normal tissue DNA.


86

Table 9: p53 polymorphisms and overexpression in ovarian carcinomas

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Case #

Histo-

SSCP

p53

Exon/

Codon/

DNA

Amino Acid

Polymor-

Immuno-

 

patho-

Result

Mut

Intron

Nucleotide

Sequence

Sequence

phism Type

staining

 

logy

 

 

 

#

 

 

 

p53

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1749

Undiff

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

1751*

Ser

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

1754

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

2341

Ser

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

2353

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

2364

Ser

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

2653*

Undiff

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

2660*

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3348

Endo

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3350*

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3351

Ser

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3355

Endo

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3357

Endo

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3359

Endo

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3362

Undiff

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3365

Endo

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3370*

Muc

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3376*

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3377*

Ser

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3379*

Ser

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3381*

Endo

+

-

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

-

3385*

Endo

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3402*

Endo

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

3403

Undiff

+

+

Intron 3

nt 11951

gggtcgggaggtccag

-

16 bp repeat

+

 

 

 

 

 

 

 

 

 

 

3379*

Ser

+

+

Exon 4

36

CCG to CCA

Pro / Pro

known PM

+

1751*

Ser

+

+

Exon 4

72

CGC to CCC

Arg to Pro

known PM

+

3350*

Ser

+

-

Exon 4

72

CGC to CCC

Arg to Pro

known PM

-

3354

Ser

+

+

Exon 4

72

CGC to CCC

Arg to Pro

known PM

+

3370*

Muc

+

-

Exon 4

72

CGC to CCC

Arg to Pro

known PM

-

3376*

Ser

+

-

Exon 4

72

CGC to CCC

Arg to Pro

known PM

-

3377*

Ser

+

-

Exon 4

72

CGC to CCC

Arg to Pro

known PM

-

3379*

Ser

+

+

Exon 4

72

CGC to CCC

Arg to Pro

known PM

+

3381*

Endo

+

-

Exon 4

72

CGC to CCC

Arg to Pro

known PM

-

3385*

Endo

+

+

Exon 4

72

CGC to CCC

Arg to Pro

known PM

+

 

 

 

 

 

 

 

 

 

 

2340

Ser

+

-

Exon 6

213

CGA to CGG

Arg / Arg

known PM

-

808

Clear

-

-

Exon 6

224

GAG to GAA

Glu / Glu

possible PM

-

 

 

 

 

 

 

 

 

 

 

3529

Ser

+

+

Intron 6

nt 13964

gcgca to gccca

-

possible PM

+

3390

Ser

+

-

Intron 6

nt 13964

gcgca to gccca

-

possible PM

-

 

 

 

 

 

 

 

 

 

 

2705

Mixed

+

-

Exon 7

231

ACC to ACA

Thr / Thr

possible PM

+

 

 

 

 

 

 

 

 

 

 

2653*

Undiff

+

-

Intron 10

nt 17708

ctact to cttct

-

possible PM

+

3402*

Endo

+

+

Intron 10

nt 17708

ctact to cttct

-

possible PM

+

 

 

 

 

 

 

 

 

 

 

2660*

Ser

+

-

Intron 10

nt 18550

ccctc to ccttc

-

possible PM

+

3344

Endo

+

+

Intron 10

nt 18550

ccctc to ccttc

-

possible PM

+

3394

Undiff

+

+

Intron 10

nt 18550

ccctc to ccttc

-

possible PM

+

3464

Ser

+

-

Intron 10

nt 18550

ccctc to ccttc

-

possible PM

+

 

 

 

 

 

 

 

 

 

 

* multiple p53 polymorphisms


87

Table 10: Normal tissue sequencing results for p53 polymorphisms

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Case #

Exon/

Codon/

Tumor

Tumor

Nl tissue

Nl tissue

Immuno-

 

Intron

Nucleotide

DNA

Amino Acid

DNA

Amino Acid

staining

 

 

#

Sequence

Sequence

Sequence

Sequence

p53

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3379

Exon 4

36

CCG to CCA

Pro / Pro

not available

-

+

1751

Exon 4

72

CGC to CCC

Arg to Pro

CGC

Arg

+

3350

Exon 4

72

CGC to CCC

Arg to Pro

CCC

Pro

-

3354

Exon 4

72

CGC to CCC

Arg to Pro

CGC

Arg

+

3370

Exon 4

72

CGC >> CCC*

Arg >> Pro

CGC

Arg

-

3376

Exon 4

72

CGC to CCC

Arg to Pro

not available

-

-

3377

Exon 4

72

CGC to CCC

Arg to Pro

CCC

Pro

-

3379

Exon 4

72

CGC / CCC

Arg / Pro

CGC

Arg

+

3381

Exon 4

72

CGC >> CCC

Arg >> Pro

CGC

Arg

-

3385

Exon 4

72

CGC to CCC

Arg to Pro

CCC

Pro

+

 

 

 

 

 

 

 

 

2340

Exon 6

213

CGA to CGG

Arg / Arg

not available

-

-

808

Exon 6

224

GAG to GAA

Glu / Glu

not available

-

-

 

 

 

 

 

 

 

 

3529

Intron 6

nt 13964

gcgca to gccca

-

not available

-

+

3390

Intron 6

nt 13964

gcgca to gccca

-

not available

-

-

 

 

 

 

 

 

 

 

2705

Exon 7

231

ACC to ACA

Thr / Thr

not available

-

+

 

 

 

 

 

 

 

 

2653

Intron 10

nt 17708

ctact to cttct

-

ctact to cttct

-

+

3402

Intron 10

nt 17708

ctact to cttct

-

not available

-

+

 

 

 

 

 

 

 

 

2660

Intron 10

nt 18550

ccctc to ccttc

-

ccctc to ccttc

-

+

3344

Intron 10

nt 18550

ccctc to ccttc

-

not available

-

+

3394

Intron 10

nt 18550

ccctc to ccttc

-

ccctc to ccttc

-

+

3464

Intron 10

nt 18550

ccctc to ccttc

-

ccctc to ccttc

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

* base G stronger than C


88

tissues to show a small proportion of cells with p53 immunostaining, usually not more than a few percent. This cut-off value of 10% had shown the best sensitivity and specificity for p53 mutations in our previous study (Wen et al. 1999). 22 (12%) cases had immunostaining <10% and 46 (26%) cases had no immunostaining of tumor cell nuclei and, together were considered to have normal p53 expression. None of the ovarian carcinomas had specific cytoplasmic staining. p53 immunostaining was not observed in the benign tissue of ovarian carcinomas including fibrous connective tissue, blood vessels and inflammatory cells. Seventy-five out of 178 (42%) cases were found to have high overexpression with a percentage of positively stained nuclei of more than 78%. This cut-point had been determined to be optimal as a predictor of clinical outcome in the statistical analysis of survival.

4.1.5 Comparison of p53 overexpression with p53 mutations

p53 overexpression detected by immunohistochemical staining was signifi-cantly correlated with p53 mutations (p<0.001) detected by DNA sequence analysis. The percentage of cases with overexpression was 81% in ovarian carcinomas with p53 mutation and 38% in cases with wildtype p53. The percentage of cases with p53 immunostaining varied according to the type of mutation. The correlation between protein overexpression and mutations was mainly due to the high proportion of missense mutations. 67 out of 71 (94%) of the cases with missense mutations showed p53 overexpression, while immunostaining ge10% was only seen in 1/7 (14%) of the cases with nonsense mutations, 8/13 (62%) of the cases with insertions or deletions and 4/8 (50%) of the cases with splice site mutations (Fig. 18 a,b).

Overall only 46% (13/28) of the ovarian carcinomas with nonmissense mutations had p53 overexpression, which is not significantly different from the frequency of overexpression in 30/79 tumors with wildtype p53 sequence (38%) (p=0.43). Nineteen percent (19/99) of the cases with p53 mutations failed to show immunostaining, which is largely caused by nonmissense mutations.

4.1.6 p53 polymorphisms and p53 protein overexpression

Ovarian cancer cases with normal p53 expression respectively p53 overexpression were distributed approximately equally among cases with exon 4 polymorphism (p=0.27) and intron 3 polymorphisms (p=0.41) (Table 11). Interestingly, though, out of eight cases with intron 6 or intron 10 sequence alterations, seven cases had p53 overexpression which is close to reaching statistical significance (p=0.086) (Table 11).


89

Fig. 18a: p53 protein overexpression according to type of p53 mutation.
Overexpression is found in a high percentage of cases with missense mutations,
while the frequency of overexpression in nonmissense mutations is overall only 46%.

Fig. 18b: p53 protein overexpression according to p53 mutation or wildtype p53.
A high proportion of cases (38%) shows p53 protein accumulation (immunostaining in
ge10% of the nuclei) despite p53 wildtype sequence.


90

Table 11: p53 protein overexpression in ovarian cancer cases with p53 polymorphisms

 

 

 

 

 

p53 Polymorphism

p53 Over-
expression
(%)

Normal p53
Expression
n (%)

Total
n (%)

p-value

 

 

 

 

 

 

 

 

 

 

Exon 4 Codon 72

 

 

 

 

Polymorphism CGC Arg rarrCCC Pro

4 (44)

5(56)

9 (5)

 

Normal Sequence CGC Arg

106 (63)

63 (37)

169 (95)

0.27

 

 

 

 

 

Intron 3

 

 

 

 

16 bp repeat Polymorphism

14 (54)

12 (46)

26 (15)

 

Normal Intron 3 Sequence

96 (63)

56 (37)

152 (85)

0.37

 

 

 

 

 

Intron 6 / Intron 10

 

 

 

 

Polymorphism

8 (89)

1 (11)

9 (5)

 

Normal Sequence

102 (60)

67 (40)

169 (95)

0.086

 

 

 

 

 

 

 

 

 

 

Total

110 (62)

68 (38)

178 (100)

 

4.1.7 Correlation of p53 alterations with histopathological and clinical data

p53 mutations were significantly more frequent in ovarian carcinomas with advanced FIGO stages III and IV (p<0.001), poor differentiation (p<0.001), residual tumor burden after debulking surgery (p<0.04), DNA-aneuploidy (p=0.003) and high S-phase-fraction as measured by image cytometry (p<0.001). Patients with p53 mutations also had a higher median age than patients with p53 wildtype (p=0.041). Overexpression of the p53 protein was correlated only with low grade of differentiation (p<0.001) and high S-phase-fraction (p<0.001) (Table 12).

4.1.8 p53 alterations and response to chemotherapy

Patients with p53 overexpressing tumors were significantly more often resistant or refractory to a platinum based chemotherapy (61%) than patients with non-overexpressing tumors (22%) (p=0.001) (Table 13, Fig. 19 A-D).


91

Table 12. Clinico-pathologic characteristics of patients with epithelial ovarian cancer

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

p53 Mutation Analysis

p53 Protein Expression

p53 Alterations

 

 

 

 

 

 

 

 

 

 

 

 

Characteristic

p53

p53

P

Overex-

normal

P

p53

normal

P

 

 

Mutation

Wildtype

value*

pression

Expression

value*

Alteration

p53

value*

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Patients - no. (%)

99 (56)

79 (44)

 

110 (62)

68 (38)

 

132 (74)

46 (26)

 

 

 

 

 

 

 

 

 

 

 

 

 

Median age - yrs

57

53

0.041**

57

52

0.06**

57

52

0.056**

 

 

 

 

 

 

 

 

 

 

 

 

Histological Subtype - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Serous

60 (61)

51 (65)

0.08

67 (61)

44 (65)

0.62

79 (60)

32 (70)

0.4

 

Mucinous

2 (2)

6 (8)

 

4 (4)

4 (6)

 

4 (3)

4 (9)

 

 

Endometriod

18 (18)

12 (15)

 

20 (18)

10 (15)

 

24 (18)

6 (13)

 

 

Clear Cell

1 (1)

5 (6)

 

2 (2)

4 (6)

 

4 (3)

2 (4)

 

 

Brenner

1 (1)

0

 

1 (1)

0

 

1 (1)

0

 

 

Undifferentiated

8 (8)

2 (3)

 

7 (6)

3 (4)

 

9 (7)

1 (2)

 

 

Mixed Epithelial

7 (7)

3 (4)

 

7 (6)

3 (4)

 

9 (7)

1 (2)

 

 

Unclassified

2 (2)

0

 

2 (2)

0

 

2 (2)

0

 

 

 

 

 

 

 

 

 

 

 

 

 

Clinical stage - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FIGO I

11 (11)

25 (32)

0.001

18 (17)

18 (28)

0.11

21 (16)

15 (34)

0.005

 

FIGO II

3 (3)

5 (7)

 

4 (4)

4 (6)

 

5 (4)

2 (7)

 

 

FIGO III

67 (70)

39 (51)

 

72 (67)

34 (52)

 

83 (64)

23 (52)

 

 

FIGO IV

15 (16)

8 (10)

 

14 (13)

9 (14)

 

20 (16)

3 (7)

 

 

 

 

 

 

 

 

 

 

 

 

 

Grade of Differentiation - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

(well)

7 (7)

21 (27)

< 0.001

8 (7)

20 (29)

< 0.001

11 (8)

17 (37)

< 0.001

 

II

(moderate)

25 (25)

27 (34)

 

32 (29)

20 (29)

 

36 (27)

16 (35)

 

 

III

(poor)

67 (68)

31 (39)

 

70 (64)

28 (41)

 

85 (64)

13 (28)

 

 

 

 

 

 

 

 

 

 

 

 

 

Residual Tumor - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No residual tumor

32 (39)

39 (56)

0.04

36 (40)

35 (56)

0.062

46 (41)

25 (60)

0.044

 

< 2 cm

29 (35)

19 (27)

 

32 (35)

16 (26)

 

37 (33)

11 (26)

 

 

> 2 cm

22 (27)

12 (17)

 

23 (25)

11 (18)

 

28 (25)

6 (14)

 

 

 

 

 

 

 

 

 

 

 

 

 

Lymph nodes - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Negative

9 (11)

15 (22)

0.2

13 (15)

11 (17)

0.5

16 (15)

8 (19)

0.24

 

Positive

20 (25)

17 (25)

 

20 (24)

17 (27)

 

24 (23)

13 (30)

 

 

Not resected

50 (63)

37 (54)

 

52 (61)

35 (56)

 

65 (62)

22 (51)

 

 

 

 

 

 

 

 

 

 

 

 

 

DNA-ploidy - no. (%) (n=102)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Diploid

21 (39)

33 (69)

0.003

28 (48)

26 (59)

0.28

31 (44)

23 (74)

0.004

 

Aneuploid

33 (61)

15 (31)

 

30 (52)

18 (41)

 

40 (56)

8 (26)

 

 

 

 

 

 

 

 

 

 

 

 

 

S-Phase-Fraction - no. (%) (n=102)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

low ( < 5%)

4 (7)

22 (46)

< 0.001

6 (10)

20 (45)

< 0.001

9 (13)

17 (55)

< 0.001

 

intermediate ( 5% - 14.4%)

26 (48)

21 (44)

 

29 (50)

18 (41)

 

35 (49)

12 (39)

 

 

high ( > 14.5%)

24 (44)

5 (10)

 

23 (40)

6 (14)

 

27 (38)

2 (6)

 

 

 

 

 

 

 

 

 

 

 

 

 

Treatment - no. (%) ***

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cisplatin - Cyclophosphamide

27 (32)

27 (39)

0.43

34 (37)

20 (33)

0.55

39 (35)

15 (38)

0.98

 

Carboplatin - Cyclophosphamide

13 (16)

7 (10)

 

13 (14)

7 (12)

 

15 (14)

5 (13)

 

 

Other regimens

25 (31)

16 (23)

 

26 (29)

15 (25)

 

31 (28)

10 (25)

 

 

No chemotherapy

17 (21)

19 (28)

 

18 (20)

18 (30)

 

26 (23)

10 (25)

 

 

No information - no.

17

10

 

19

8

 

21

6

 

 

 

 

 

 

 

 

 

 

 

 

* Chi-Square test for two categories, Mantel-Haenszel test for three or more categories
** Wilcoxon test
*** Chemotherapy regimens are described in the methods section


92

Among patients with p53 mutations, 56% (22/39) as opposed to only 35% (12/34) of the patients with wildtype p53 were resistant or refractory to platinum based chemotherapy, but the result did not achieve statistical significance (p=0.071). However the response to chemotherapy was correlated with the type of mutation. If only patients with missense mutations were evaluated, only 10/29 (34%) were platinum sensitive as compared to 29/44 (66%) of patients with p53 wildtype or nonmissense mutations (p=0.008) (Table 13). Overall, time to progression since chemotherapy was significantly shorter for patients with p53 alterations than patients with normal p53 (p=0.037) (Fig. 19).

Table 13: Response to platinum-based chemotherapy in correlation to p53 alterations

 

Immunostaining

 

p53 Sequence

 

Type of Mutation

 

 

 

 

 

 

 

 

Response to
Chemotherapy

Neg

n (%)

Pos

n (%)

WT

n (%)

MUT

n (%)

WT/NonMissense n(%)

Missense


n(%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sensitive

21 (78)

18 (39)

22 (65)

17 (44)

29 (66)

10 (34)

 

 

 

 

 

 

 

Resistant / Refractory

6 (22)

28 (61)

12 (35)

22 (56)

15 (34)

19 (66)

 

 

 

 

 

 

 

p-value

 

0.001

 

0.071

 

0.008

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Total (100%)

27

46

34

39

44

29

Interestingly, out of 5 patients in our study who were treated with Taxol/Carboplatin, the only patient (case #3400) who was sensitive to treatment and had no evidence of disease 28 months after diagnosis of a poorly differentiated FIGO III ovarian carcinoma, had high overexpression of p53 protein and a missense mutation in exon 5.

4.1.9 p53 alterations as a predictor of time to progression and overall survival

Ovarian cancer patients with p53 mutations had a significantly shorter time to progression (p=0.029) and overall survival (p=0.014) than patients with wildtype p53 (Fig. 20A, Table 14, 15). In 61/99 (62%) tumors, the p53 mutation was in an evolutionary highly conserved domain. If the clinical outcome of these was


93

Fig. 19 A-B: Estimated probability of not progressing in epithelial ovarian cancer
patients (n=74), who received platinum-based chemotherapy.
(A) Ovarian carcinomas with p53 mutations versus wildtype p53. (B) Ovarian carcinomas with p53 proteinoverexpression versus normal p53 expression.


94

Fig. 19 C-D:: Estimated probability of (C) not progressing, and (D) overall
survival respectively in epithelial ovarian cancer patients (n=74), who received
platinum-based chemotherapy.
Ovarian cancer with p53 alterations (mutation or
overexpression or both) versus normal p53 status.


95

compared to patients with either wild-type p53 or a mutation in a nonconserved domain, the difference in time to progression (p=0.010) and overall survival was even more significant (p=0.007) (Fig. 20B). Similar results were seen for p53 mutations in DNA interacting regions (Fig. 20C). Furthermore, time to progression and overall survival were better in patients with normal p53 expression as opposed to p53 overexpression, but the results did not reach statistical significance (p=0.071, p=0.056 logrank) (Fig. 20D, Table 14, 15). In ovarian cancer cases with high p53 overexpression (>78%) though, time to progression and overall survival were significantly shorter than in cases with low overexpression or normal p53 expression (p=0.043, p=0.020 logrank). Overall, the most favourable prognosis in terms of overall survival was seen in patients who had wildtype p53 sequence and normal p53 expression as opposed to those, who had either one or both alterations of the p53 gene and p53 protein expression (p=0.007) (Fig. 20E,F).

4.1.10 Univariate and multivariable analysis of prognostic factors

Univariate analysis of time to progression identified p53 mutations, p53 alterations, residual disease, FIGO stage, histologic grade, age, lymph node status, and S-phase-fraction as prognostic factors. Multivariable analysis of time to progression identified only FIGO stage (p=0.001), residual disease (p=0.005) and age (p=0.032) as independent prognostic factors (Table 14).

Univariate analysis of overall survival identified residual disease, FIGO stage, age, grade, lymph node status and S-phase-fraction as prognostic factors besides p53 mutation and p53 alteration (Table 15). In multivariable analysis though, only residual disease (p<0.001), FIGO stage (p=0.001), grade (p=0.009), and patient age (p=0.01) were shown to be independent predictors of survival (Table 15).


96

Fig. 20 A-B: Estimated probability of overall survival in patients with epithelial
ovarian cancer (n=178) according to p53 alterations.
(A) p53 mutations versus
wildtype p53. (B) p53 mutations in evolutionary highly conserved domains versus
wildtype p53 or mutations in non-conserved domains.


97

Fig. 20 C-D: Estimated probability of overall survival in epithelial ovarian
patients (n=178) according to p53 alterations.
(C) p53 mutations in DNA-
interacting regions versus wildtype p53 or mutations in scaffolding regions of the p53
protein. (D) p53 protein overexpression versus normal expression of the p53 protein.


98

Fig. 20 E-F: Estimated probability of (E) not progressing, and (F) overall survival
respectively in epithelial ovarian cancer patients (n=178).
Ovarian cancer with p53
alterations (p53 mutation or p53 overexpression or both) versus normal p53 status.


99

Table 14: Multivariable proportional hazards Cox regression analysis for the identification of independent prognostic factors in time to progression of patients with ovarian carcinoma.

 

 

 

 

 

 

 

 

 

Variables

 

Univariate Analysis

 

 

Multivariable Analysis

 

 

 

 

 

 

 

 

 

 

Variables in the

 

Unadjusted

 

 

 

Adjusted

 

 

Baseline Model *

 

Relative Risk

(95% CI)

p-value[1]

 

Relative Risk

(95%CI)

p-value[2]

FIGO

 

 

 

< 0.001

 

 

 

0,001

I + II

 

1,00

 

 

 

1,00

 

 

III + IV

 

6,08

(3.16, 11.71)

 

 

3,76

(1.70, 8.32)

 

Residual Disease

 

 

 

< 0.001

 

 

 

0,005

None

 

1,00

 

 

 

1,00

 

 

Any

 

3,89

(2.50, 6.06)

 

 

1,98

(1.19, 3.27)

 

Histologic Grade

 

 

 

< 0.001

 

 

 

0,44

Well Differentiated

 

1,00

 

 

 

1,00

 

 

Moderately

 

1,95

(0.99, 3.85)

 

 

1,23

(0.54, 2.81)

 

Poorly

 

3,00

(1.58, 5.68)

 

 

1,52

(0.70, 3.34)

 

The p53 Variables

 

Unadjusted

 

 

 

Adjusted

 

 

 

 

Relative Risk

(95% CI)

p-value[1]

 

Relative Risk

(95%CI)

p-value[2]

Immunostaining

 

 

 

0,071

 

 

 

0,67

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,42

(0.96, 2.08)

 

 

1,10

(0.71, 1.69)

 

Mutation

 

 

 

0,029

 

 

 

0,82

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,51

(1.04, 2.20)

 

 

0,95

(0.62, 1.47)

 

Alteration

 

 

 

0,030

 

 

 

0,59

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,59

(1.03, 2.47)

 

 

1,15

(0.70, 1.90)

 

Conserved Regions

 

 

 

0,010

 

 

 

0,47

Non Conserved/ WT p53

 

1,00

 

 

 

1,00

 

 

Cons. Region II, III, IV, V

 

1,66

(1.14, 2.42)

 

 

1,17

(0.77, 1.79)

 

% p53 IM (continuous)

 

 

 

0,13

 

 

 

0,40

0

 

1,00

 

 

 

1,00

 

 

10

 

1,03

(0.99, 1.08)

 

 

0,98

(0.93, 1.03)

 

20

 

1,07

(0.98, 1.16)

 

 

0,96

(0.87, 1.05)

 

50

 

1,18

(0.95, 1.46)

 

 

0,90

(0.71, 1.14)

 

Other Variables

 

Unadjusted

 

 

 

Adjusted

 

 

of Interest

 

Relative Risk

(95% CI)

p-value[1]

 

Relative Risk

(95%CI)

p-value[2]

Age [0]

 

 

 

0,019

 

 

 

0,032

50

 

1,00

 

 

 

1,00

 

 

60

 

1,19

(1.03, 1.38)

 

 

1,21

(1.02, 1.44)

 

70

 

1,42

(1.06, 1.90)

 

 

1,46

(1.04, 2.06)

 

Metastasis

 

 

 

0,057

 

 

 

0,63

None

 

1,00

 

 

 

1,00

 

 

Any

 

1,82

(1.02, 3.23)

 

 

1,16

(0.63, 2.15)

 

Lymph Node Status

 

 

 

< 0.001

 

 

 

0,31

Free of Tumor

 

1,00

 

 

 

1,00

 

 

Infiltrated

 

4,84

(2.00, 11.73)

 

 

1,91

(0.76, 4.76)

 

Not Resected

 

3,57

(1.54, 8.29)

 

 

1,84

(0.76, 4.44)

 

Histology

 

 

 

0,18

 

 

 

0,37

Serous

 

1,00

 

 

 

1,00

 

 

Other

 

0,76

(0.51, 1.14)

 

 

1,23

(0.78, 1.95)

 

S-Phase

 

 

 

0,006

 

 

 

0,54

0.0 - 4.9%

 

1,00

 

 

 

1,00

 

 

5.0 - 14.4%

 

1,76

(0.83, 3.77)

 

 

1,44

(0.54, 3.84)

 

14.5% +

 

3,29

(1.51, 7.18)

 

 

1,79

(0.61, 5.25)

 

Ploidy

 

 

 

 

1,00

 

 

 

0,45

Diploid

 

1,00

 

 

 

1,00

 

 

Tetraploid / Aneuploid

 

1,00

(0.60, 1.68)

 

 

0,79

(0.43, 1.46)

 

* Based on a previous analysis, FIGO, residual disease, and histologic grade were selected for the baseline model
[0] Age, in years, fit as a continuous variable.
[1] P-values based on the likelihood ratio of the univariate model with no other variables.
[2] P-values based on the likelihood ratio with the variables included in the baseline model:
FIGO, residual disease, and histologic grade.


100

Table 15: Multivariable proportional hazards Cox regression analysis for the identification of independent prognostic factors in overall
survival of patients with ovarian carcinoma.

 

 

 

 

 

 

 

 

 

Variables

 

Univariate Analysis

 

 

Multivariable Analysis

 

 

 

 

 

 

 

 

 

 

Variables in the

Baseline Model *

 

Unadjusted

Relative Risk

(95% CI)

p-value[1]

 

Adjusted

Relative Risk

(95%CI)

p-value[2]

FIGO

 

 

 

< 0.001

 

 

 

0,002

I + II

 

1,00

 

 

 

1,00

 

 

III + IV

 

6,98

(3.51, 13.87)

 

 

3,59

(1.49, 8.66)

 

Residual Disease

 

 

 

< 0.001

 

 

 

< 0.001

None

 

1,00

 

 

 

1,00

 

 

Any

 

4,77

(3.00, 7.58)

 

 

2,44

(1.46, 4.09)

 

Histologic Grade

 

 

 

< 0.001

 

 

 

0.009

Well Differentiated

 

1,00

 

 

 

1,00

 

 

Moderately

 

3,99

(1.67, 9.50)

 

 

3,26

(1.13, 9.43)

 

Poorly

 

5,95

(2.58, 13.71)

 

 

3,87

(1.37, 10.88)

 

The p53 Variables

 

Unadjusted

 

 

 

Adjusted

 

 

 

 

 

Relative Risk

(95% CI)

p-value[1]

 

Relative Risk

(95%CI)

p-value[2]

Immunostaining

 

 

 

0.056

 

 

 

0.73

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,45

(0.98, 2.14)

 

 

1,08

(0.70, 1.67)

 

Mutation

 

 

 

0.014

 

 

 

0.49

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,59

(1.09, 2.33)

 

 

0,86

(0.55, 1.33)

 

Alteration

 

 

 

0.005

 

 

 

0.47

No

 

1,00

 

 

 

1,00

 

 

Yes

 

1,88

(1.18, 3.00)

 

 

1,21

(0.72, 2.05)

 

Conserved Regions

 

 

 

0.007

 

 

 

0.8

Non Conserved / WT

 

1,00

 

 

 

1,00

 

 

Cons. Regions II,III,IV,V

 

1,70

(1.17, 2.47)

 

 

1,06

(0.69, 1.61)

 

Im p53 (continuous)

 

 

 

0.056

 

 

 

0.64

0

 

1,00

 

 

 

1,00

 

 

10

 

1,04

(1.00, 1.09)

 

 

0,99

(0.94, 1.04)

 

20

 

1,09

(1.00, 1.19)

 

 

0,98

(0.89, 1.07)

 

50

 

1,23

(0.99, 1.53)

 

 

0,95

(0.75, 1.20)

 

Other Variables of Interest

 

Unadjusted Relative Risk

(95% CI)

p-value[1]

 

Unadjusted Relative Risk

(95%CI)

p-value[2]

Age [0]

 

 

 

0.003

 

 

 

0.005

50

 

1,00

 

 

 

1,00

 

 

60

 

1,26

(1.08, 1.46)

 

 

1,29

(1.08, 1.55)

 

70

 

1,58

(1.17, 2.14)

 

 

1,67

(1.17, 2.40)

 

Metastasis

 

 

 

0.062

 

 

 

0.6

None

 

1,00

 

 

 

1,00

 

 

Any

 

1,77

(1.01, 3.10)

 

 

1,18

(0.64, 2.17)

 

Lymph Node Status

 

 

 

0,001

 

 

 

0,54

Negative

 

1,00

 

 

 

1,00

 

 

Positive

 

4,20

(1.82, 9.68)

 

 

1,54

(0.64, 3.71)

 

Not Resected

 

3,13

(1.42, 6.87)

 

 

1,26

(0.54, 2.94)

 

Histology

 

 

 

0.021

 

 

 

0.9

Serous

 

1,00

 

 

 

1,00

 

 

Other

 

0,78

(0.52, 1.16)

 

 

1,03

(0.65, 1.64)

 

S-Phase-Fraction

 

 

 

0.001

 

 

 

0.72

0.0 - 4.9%

 

1,00

 

 

 

1,00

 

 

5.0 - 14.4%

 

2,75

(1.13, 6.68)

 

 

1,47

(0.43, 5.07)

 

14.5% +

 

4,88

(1.95, 12.17)

 

 

1,68

(0.46, 6.19)

 

DNA-Ploidy

 

 

 

0.24

 

 

 

0.37

Diploid

 

1,00

 

 

 

1,00

 

 

Tetraploid / Aneuploid

 

1,38

(0.80, 2.37)

 

 

1,33

(0.71, 2.51)

 

* Based on a previous analysis, FIGO, residual disease, and histologic grade were selected for the baseline model
[0] Age, in years, fit as a continuous variable.
[1] P-values based on the likelihood ratio of the univariate model with no other variables.
[2] P-values based on the likelihood ratio with the variables included in the baseline model:
FIGO, residual disease, and histologic grade.


101

4.2 Alterations of the mdm2 gene in ovarian cancer

4.2.1 Expression and absence of amplification of mdm2 in ovarian carcinomas

Fifty-six ovarian cancer cases were analyzed by Southern hybridization for mdm2 DNA-amplification. The osteosarcoma cell line SA1 which is known to have amplification and overexpression of mdm2 was used as a control (Fig. 21). Fifty-two ovarian cancer cases did not show amplification of the mdm2 gene, independently of p53 mutations and/or overexpression. In four cases, the result could not be interpreted. All cases had been previously sequenced and analyzed for p53 protein overexpression. In 21 of these cases, p53 wildtype had previously been found. Thirteen of these 21 cases showed p53 overexpression despite the wild type sequence. In 30 cases, p53 was mutated, 26 of which also showed protein overexpression.

Fig. 21: Southern hybridization of genomic DNA with the mdm2 cDNA probe. The sarcoma cell line SA1 which was used as a positive control shows amplification of
mdm2 DNA, while none of the ovarian cancer tissues shows DNA amplification.

mdm2-expression was also analyzed by Northern hybridization in 29 ovarian cancer cases (Fig. 22). Only the sarcoma cell line SA1 but none of the ovarian cancer cases showed overexpression of the mdm2 gene. In the sarcoma cell line as well as all ovarian cancers, three different mRNA transcripts of 7.4 kb, 5.5 kb and 2.8 kb were seen. Ten ovarian cancer cases had previously been shown to have wildtype p53, of which six overexpressed p53 as detected by immunohistochemistry. The remaining 19 cases had been identified to have a p53 mutation, 16 of which also showed protein overexpression.


102

Fig. 22: Northern hybridization of total mRNA with the mdm2 cDNA probe. The
sarcoma cell line SA1 which was used as a positive control shows overexpression of
mdm2 RNA, while none of the ovarian cancer tissues shows mdm2 overexpression.

4.2.2 mdm2 alternative and aberrant RNA splicing in ovarian carcinomas

PCR analysis of the reverse transcribed mRNA revealed the presence of the ex-pected full length (1473) bp product of mdm2 as well as other, smaller products (Fig. 23). Among 92 cases analyzed, 76 (83%) had a full length 1473 bp RT-PCR product, in some cases only weakly expressed. Only 18 cases (20%) had exclusive expression of the normal mdm2 full length cDNA, while 58 cases (63%) had both the 1473 bp product and smaller RT-PCR products (Table 16). Eight cases (9%) showed no mdm2 expression despite repeated RT-PCR analysis.

Amplification of the human beta2-Microglobulin gene, a member of the immunoglobulin gene superfamily, which is represented on the surface of nearly all cells (Güssow et al. 1987), was used as a control for integrity and equal amounts of RNA respectively cDNA. PCR analysis of the beta2-Microglobulin showed strong expression in all normal ovarian tissues and all ovarian carcinomas (Fig. 24).


103

Fig. 23: mdm2 RNA splice variants in ovarian carcinomas. Total RNA was
ana-lyzed by RT-PCR, and PCR-products were separated on a 1.2% agarose gel. Cases
in lanes 1,2,6,7,9, and 10 show splice variants of different sizes. The case in lane 3
shows no mdm2 expression. (cDNA bands on the gel contain additional 53 bp of flanking primer sequence)

Table 16: mdm2 RNA splicing in ovarian cancer as analyzed by reverse transcriptase PCR and DNA Sequencing

RT-PCR result in ovarian carcinomas

n (%)

n (%)

 

 

 

normal mdm2 only

18 (20)

 

normal mdm2 and splice variants

58 (63)

 

Normal mdm2 / mdm2-b

 

26 (28)

Normal mdm2 / mdm2-b / mdm2 221

 

1 (1)

Normal mdm2 / mdm2 221

 

8 (9)

Normal mdm2 / other splice variants

 

23 (25)

splice variants only / no normal mdm2

8 (9)

 

Mdm2-b only

 

3 (3)

Other splice variants only

 

5 (5)

no mdm2 expression

8 (9)

 

Total

92 (100)

 


104

Fig. 24: A-B: RT-PCR for ß2 microglobulin as a positive control for quality and
amount of RNA.
(A) RNA was extracted from normal ovarian tissue. All cases show
equal amplification of a 898 bp ß2M PCR product. Lane 21 shows the PCR product of
cDNA from the SA1 sarcoma cell line. Lane 22 is the negative control. (B) RT-PCR
for ß2 microglobulin expression in ovarian cancer RNA. Except for case #2341 in lane
7 all cases show approximately equal expression of the ß2 microglobulin RNA.

Out of those cases which showed an approximately 700 bp band on the agarose gel, six ovarian cancer cases, one borderline tumor, and one normal ovarian tissue were sequenced and the 654 bp splice variant was confirmed. The typical approximately 290 bp band on the agarose gel was confirmed as the 221bp splice variant in 8 ovarian carcinomas, four tumors of borderline malignancy, one cystadenoma, and four normal ovarian tissues.


105

The mdm2 splice variant which we found most frequently was a 654 base mdm2 cDNA. We found this splice variant in 38 (41%) ovarian cancer cases. It comprises 654 bp of the open reading frame and splices out exon 4-11 (Fig. 23, Table 17). The 654 bases fragment misses a large portion of the mRNA, including 90% (81 of 90 amino acids) of the 3‘ end of the p53 binding domain, and the entirety of the nuclear localization signal and acidic domain (Fig. 25). This splice variant had been named mdm2-b by Sigalas et al. (1996) and was described as a 707 bp product, since he included the primer sequence which is outside the open reading frame of mdm2 into the numbering (Sigalas et al. 1996). We named the fragment 654 bp, because we counted only those bases between the start and the stop codon, but not the primers.

Another splice variant, which has been named mdm2-a (Sigalas et al. 1996) splices out exon 4 through 9 and spans 888 basepairs of the open reading frame (Fig. 25, Table 17). This variant was seen in 4 cases (4 %). In both of these variants splicing takes place at the exon/intron boundary and the truncated RNA is in frame.

In contrast to this in a small splice variant of 221 bp, which was seen in 15 cases (16%), splicing did not take place at the exon/intron boundary. We identified three variations of this splice variant (Table 17).

We identified a total of 30 different splice variants. Fourteen of these (47%) were only seen in ovarian carcinomas, while seven were identified both in ovarian cancer and other histologies. Seven were seen only in borderline tumors or cystadenomas, and two splice variants were found exclusively in normal ovarian tissue. Most of these splice variants should be classified as aberrant splicing because they splice out parts of exon sequences (Table 17).

In only three splice variants of 888 bp (mdm2-a), 654 bp (mdm2-b) and 613 bp, donor and acceptor splice site were found at exon-intron boundaries. The mdm2-a splice variant splices out exon 4 through exon 9 and the mdm2-b splice variant splices out exon 4 through exon 11. In the 613 bp splice variant the entire exon 5 is spliced out the exon/intron boundaries as well as part of exon 9, exon 10, exon 11, and part of exon 12 (Table 17). In the splice variants of 446 bp, 397 bp, 391 bp, 365 bp, 357 bp, and 351 bp length, the donor splice site is at the exon/intron-boundary of exon 4, 5 or 6, while the acceptor splice site is an internal splice site in exon 12. The splice variants of 391 bp and 357 bp splice out exon 5-11 and part of exon 12, the splice variants of 397 bp and 351 bp splice out exon 6-11 and part of exon 12, and the splice variants of 446 bp and 365 bp splice out exon 7-11 and part of exon 12.

All other splice variants would be considered aberrant splicing since they have a donor splice site internally in exon 3-10 and an acceptor splice site internally in exon 12. Except for the mdm2-a splice variant, exon 11 is generally spliced out.


106

Fig. 25: Loss of functional regions in mdm2 splice variants in comparison to the full length mdm2 gene Above, functional domains of the MDM2 protein with corresponding amino acid residues are shown (according to Freedman et al. 1999). Below, splice variants of the mdm2 gene are shown. Black lines indicate the missing part of the sequence.


107-108

Table 17: mdm2 RNA splice variants in ovarian cancer identified by RT-PCR and DNA sequencing. The table shows sizes of splice variants, splice sites and tissues in which the splice variant was identified. (First nucleotide of coding sequence is #312, last nucleotide of coding sequence is #1784). The length of the splice variants includes only the coding sequence but not the flanking primer sequence. (> indicates exon/intron boundary and < indicates intron/exon boundary)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Size of

p53

Donor

Donor

Splice

site

sequence

Acceptor

Acceptor

Overlap

Frame

Splice variant identified in

splice

binding

splice site

splice site

 

 

 

splice site

splice site

 

 

 

 

 

 

variant

site

nucleotide #

exon

Donor

 

Acceptor

exon

(nucleotide #)

 

 

Ov

LMP

Cystade-

NL

 

 

 

 

 

 

 

 

 

 

 

Ca

 

noma

Ovary

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

888

partial

392

ex 3 >

GAG ACC CTG

-----

GAT CTT GAT

< ex 10

978

G

in frame

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

721

complete

1007

cryptic ex 10

AGT GAA CAT TCA

-----

A ATG ATT

cryptic ex 12

1760

ATTCA

shift (+A..)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

654

Partial

392

ex 3 >

GAG ACC CTG

-----

GAC TAT TGG

< ex 12

1212

G

in frame

+

+

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

613

Partial

467

ex 4 >

ATG AAA GAG

-----

G AAA ATA

< ex 6

602

AGG

shift (+G)

-

-

+

-

 

 

and

and

and

 

 

 

 

 

 

 

 

 

 

 

Partial

856

cryptic ex 9

CGA CAA AGA A

-----

GAC AAA GAA

cryptic ex 12

1581

GACAAAGAA

shift (+A)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

571

Complete

879

cryptic ex 9

GAT AGT ATT T

-----

C CCC

cryptic ex 12

1781

TATTT

shift (+T,+C)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

565

Complete

807

cryptic ex 8

AGA GCA ATT A

-----

AAG AAA AGG

cryptic ex 12

1716

TA

shift (+A)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

541

Complete

827

cryptic ex 9

GAA AAT TCA

-----

A ATG ATT GTG

cryptic ex 12

1760

AATTCA

shift (+A)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

533

Partial

407

cryptic ex 4

ACC AAA GCC A

-----

AA CTG GAA

cryptic ex 12

1348

AAAGCCA

in frame

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

520

partial

596

cryptic ex 5

CTC TGT GAA GAG

-----

TTT GAA AGG

cryptic ex 12

1550

TGTGAAGAG

in frame

-

-

+

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

461

partial

510

cryptic ex 5

AAA CGA TTA T

-----

T TAT AGC AGC

cryptic ex 12

1523

ATTAT

shift (+TT)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

446

complete

650

ex 6 >

AAT CAG CAG

-----

GA CAT CTT ATG

cryptic ex 12

1678

CAGGA

shift (+GA)

-

+

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

442

partial

453

cryptic ex 4

AAA GAC ACT T

-----

GAA GAC TAT

cryptic ex 12

1485

T

shift (+T)

-

-

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

421

complete

627

cryptic ex 6

ATC TAC AGG A

-----

CAT CTT ATG

cryptic ex 12

1680

ACAGGA

shift (+A)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

421

complete

683

cryptic ex 7

GTG AGT GAG

-----

C CCA GTA TGT

cryptic ex 12

1736

G

shift (+C)

-

-

+

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

397

partial

601

ex 5 >

GAG CAC AG

-----

GA CAT CTT ATG

cryptic ex 12

1678

ACAGGA

shift (+GA)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

391

partial

467

ex 4 >

ATG AAA GAG

-----

TTT GAA GGG

cryptic ex 12

1551

TGAAAGAG

in frame

+

-

+

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

378

partial

408

cryptic ex 4

CCA AAG CCA

-----

TCA ACT TCT

cryptic ex 12

1504

AGCCAT

in frame

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

365

complete

651

ex 6 >

AAT CAG CAG G

-----

A ATG ATT GTG

cryptic ex 12

1760-

AAT

shift (+GA)

-

-

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

357

partial

467

ex 4 >

ATG AAA GAG

-----

AGT GTG GAA

cryptic ex 12

1584

AAGAG

in frame

-

+

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

351

partial

601

ex 5 >

GAG CAC AG

-----

G AAT AAG CCC

cryptic ex 12

1724

AGGAA

in frame

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

334

partial

597

cryptic ex 5

GTG AAA GAG C

 

CCA GTA TGT AGA

cryptic ex 12

1737

GC

(shift+C)

-

+

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

282

partial

522

cryptic ex 5

GAT GAG AAG C

-----

TA AAGAAA

cryptic ex 12

1714

AGAAGC

in frame

-

+

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

279

partial

514

cryptic ex 5

CGA TTA TAT GA

-----

G AAG CTA AAG

cryptic ex 12

1712

A

in frame

+

-

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

267

partial

447

cryptic ex 4

GCA CAA AAA G

-----

GT TGC ATT AGA

cryptic ex 12

1654

G

in frame

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

266

partial

410

cryptic ex 4

AAG CCA TTG

-----

AA CCT TGT

cryptic ex 12

1618

AGCCATTG

shift (+AA)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

259

partial

551

cryptic ex 5

TGT TCA AAT GAT

-----

T GTG CTA ACT

cryptic ex 12

1766

TTCAAATGAT

shift (+T)

+

-

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

221

partial

483

cryptic ex 5

TAT CTT G

-----

C CCA GTA

cryptic ex 12

1736

TG

shift (+GC)

+

+

+

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

221

partial

484

cryptic ex 5

TAT CTT GGC

-----

CCA GTA

cryptic ex 12

1737

GCC

shift (+GC)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

221

partial

491

cryptic ex 5

TAT CTT GGC CAG TAT

-----

GT AGA CAA

cryptic ex 12

1744

CCAGTAT

shift (+GT)

+

-

-

+

 

 

484

 

or TAT CTT GG

-----

CCA GTA TGT AGA

 

1737

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

150

partial

455

cryptic ex 4

GAC ACT TAT

-----

TTC CCC

cryptic ex 12

1779

ACTTAT

in frame

+

+

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

77

absent

338

cryptic ex 3

TCT GTA CCT

-----

GC CCA GTA

cryptic ex 12

1735

CCT

shift (+GC)

+

-

-

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

52

absent

350

cryptic ex 3

GAT GGT GCT

-----

A ACT TAT

cryptic ex 12

1772

GTGCT

shift (+A)

+

-

-

+

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


109

4.2.3 Loss of p53 binding sequence in mdm2 splice variants

The binding site of the p53 protein has been suggested to be a region of approximately 84 amino-acids between aa 19-102 of the MDM2 protein (Chen et al. 1993). This corresponds to base 366-617 of the mdm2 sequence. Out of 30 splice variants which had been identified in ovarian tumors and normal ovarian tissue, only 8 splice variants (27%) contained the complete p53 binding sequence, while 20 (67%) had only a partial sequence and 2 (7%) of the splice variants were missing the entire p53 binding site (Table 18, Fig. 25). The most frequent splice variant of 654 bp length (mdm2-b) contains only 9 out of 84 amino acids of the p53 binding site, while the 221 bp splice variant contains 39 amino acids of the binding site.

Table 18: Loss of p53 binding sequence in mdm2 splice variants

p53 binding site in
mdm2 splice variants

n (%)

size of splice variant (bp)

 

 

 

Complete p53 binding site

8 (27%)

721,571, 565,541,446,421,421,365

Partial p53 binding site

20 (67%)

888,654,613,533,520,461,442,397,391,378,
357,351,334,282,279,267,266,259,221,150

Loss of p53 binding site

2 (7%)

77,52

 

 

 

 

30 (100%)

 

4.2.4 mdm2 splice sites and repeat sequences

Many of the splice variants show an interesting pattern of splicing, because the donor and acceptor sites are in regions of exact sequence homology (Fig. 26). For example in the 391 bp splice variant, which was seen in carcinomas as well as in benign tissues, the 3‘ end of the donor site ends with an eight base sequence of TGAAAGAG which is also found at the non-transcribed downstream acceptor splice site in exon 12 (Fig. 26). The acceptor splice site sequence would be tgaaagagTTTGAAAGG. (The underlined sequence indicates the exact sequence match. The uppercase letters indicate the coding bases).


110

In the 221 bp splice variant, the splice site showed sequence overlaps with minor variations in different cases which are shown in Table 17 and Fig. 26. Similar sequence matches, mostly 2 to 10 bases long, were found in the majority of the splice variants.

Fig. 26: A-D: Sequence overlaps in mdm2 RNA splice variants. (A-C) Splice site sequences of the 391 bp and 221 bp mdm2 splice variants show overlaps of several basepairs. The capital letters refer to transcribed bases that remain in the final mRNA products, while the lower case letters refer to untranscribed bases which are spliced out of the final mRNA product. The green boxes enclose the transcribed sequence at the donor site and the blue boxes enclose the transcribed sequence at the acceptor site. The yellow boxes indicate sequence homologies. D) As opposed to many of the aberrant splice variants which show overlapping sequences at the splice site, no such sequence homology is notable in the 654 bp mdm2-b splice variant which splices at exact exon/intron boundaries of exon 3 and exon 12.


111

Interestingly, the acceptor splice site of the 221 bp splice variant (nucleotide 1737) was shared by four different splice variants of 421 bp, 334 bp, 22 bp and 77 bp length (Table 17). The same phenomenon was observed for other splice variants which used the exact same sequence in exon 12 as acceptor site. The sequence match in all three cases comprises a CAGGA sequence. The 446 bp splice variant has the donor site at the exon 6/intron 6 boundary with an AAT CAG CAG gaa tca tcg sequence and splices into exon 12 at base 1678 with a ggc aaa aca gGA CAT CTT ATG sequence.

The 421 bp splice variant has a donor site in exon 6 with an ATC TAC AGg aac ttg gta sequence and splices into exon 12 at base 1678 with the same ggc aaa aca gGA CAT CTT ATG sequence. And the 397 bp splice variant has the donor site at the exon 5/intron 5 boundary with a GAG CAC AGg aaa ata tat sequence and splices into the identical acceptor site in exon 12 (Table 17). All three splice variants are out of frame. The donor sites were shared among the splice variants 888 (mdm2-a) and 654 (mdm2-b) (nucleotide 392), 613, 391, and 357 (nucleotide 467), and 397 and 351 (nucleotide 601).

Twelve of the thirty splice variants with a cDNA length between 888 and 150 bp were in frame, while all others were out of frame and therefore, either no protein or only a truncated protein will be synthesized.

4.2.5 mdm2 alterations in ovarian cystadenomas and borderline tumors

RNA from six benign cystadenomas and nine ovarian tumors of low malignant potential (LMP) was analyzed by RT-PCR for mdm2 alterations (Table 19 and 20, Fig. 27). Except for one cystadenoma which was completely lacking mdm2 expression, all other cystadenomas showed the full length mdm2 as well as splice variants (Table 19). None of the cases had only a full length mdm2 transcript. The 221 bp splice variant, which had been described for ovarian carcinomas, was seen in one case. Four other cystadenomas, though, showed splice variants which had not been found before in the ovarian cancer cases. The length of these splice variants was 613, 520, 421 and 390 basepairs. The mdm2-b splice variant of 654 bp, which is frequent in ovarian carcinomas, was not seen in any of the cystadenomas.

Among the nine borderline ovarian tumors (Table 20), only one case was found to have only the full length mdm2 cDNA and one case had no mdm2 expression. The other seven cases showed mdm2 alternative splicing, which was seen in six cases, in addition to the normal length mdm2, and in one case as the only mdm2 cDNA.


112

Fig. 27: mdm2 RNA splice variants in ovarian cystadenomas and tumors of
borderline malignancy.
Total RNA was analyzed by RT-PCR and PCR-products were
separated on a 1.2% agarose gel. Cases in lanes 1, 2, 5-8, and 10 show splice variants
of different sizes. The case in lane 3 shows no mdm2 expression despite normal
expression of the ß2-microglobulin control gene. (cDNA bands on the gel contain additional 53 bp of flanking primer sequence).

Table 19: mdm2 alternative and aberrant splicing in benign ovarian cystadenomas

Case
ID #

Diagnosis at
surgery

full length
mdm2

mdm2-b
654 bp

mdm2
221 bp

other splice
variants

 

 

 

 

 

 

2305

serous cystadenoma

+

-

+

-

2539

serous cystadenoma

-

-

-

-

2771

serous cystadenoma

+

-

-

421

3456

serous cystadenoma

+

-

-

613

3457

mucinous cystadenoma

+

-

-

391

3462

serous cystadenoma

+

-

-

520

 

 

 

 

 

 

n (%)

 

5 (83)

0 (0)

1 (17)

4 (67)


113

Table 20: mdm2 alternative and aberrant splicing in ovarian tumors of borderline malignancy

Case
ID #

Diagnosis at
surgery

full length
mdm2

mdm2-b
654 bp

mdm2
221 bp

other splice
variants

 

 

 

 

 

 

2434

serous borderline

+

-

+

-

3084

serous borderline

+

-

+

-

3453

serous borderline

+

-

-

-

3454

serous borderline

+

-

+

334, 150

3455

serous borderline

+

-

+

-

3458

mucinous borderline

+

-

-

357, 282

3459

serous borderline

-

-

-

-

3463

mucinous borderline

+

-

-

446

3465

serous borderline

-

+

+

-

 

 

 

 

 

 

n (%)

 

7(78)

1 (11)

5 (56)

3 (33)

The mdm2-b splice variant (654 bp) was seen in one (1/9) of the borderline ovarian tumors (11%). Five cases had the 221 bp splice variant, which is also frequent in ovarian carcinomas. Five splice variants of the sizes 446, 357, 334, 282 and 150 bp were found in three cases of borderline tumors. The 150 bp splice variant was also found in two ovarian carcinomas, while the other variants were only seen in the borderline tumors.

4.2.6 mdm2 alterations in normal ovarian tissue

RNA from normal ovarian tissue was extracted and analyzed for mdm2 splice products by RT-PCR (Fig. 28). The age of the patients, diagnosis at the time of surgery, menopausal status or day of menstrual cycle respectively are demonstrated in table 21. In nine cases the ovaries were removed in surgery for uterine fibroids. In one patient the ovarian tissue was snap frozen at the time of autopsy for liver failure. Two patients had ovarian cancer of the contralateral side, two patients had endometrioid cancer, and two had cervical cancer.

Only 9/20 (45%) of the normal ovarian tissues showed expression of only the full length mdm2 cDNA, while 11/20 (55%) showed alternative splice products, in


114

Fig. 28: mdm2 RNA splice variants in normal ovarian tissue. Total RNA of
ovarian tissue was analyzed by RT-PCR and PCR-products were separated on a 1.2%
agarose gel. Cases in lanes 3, 5, 7, and 8 show splice variants of different sizes. (cDNA bands on the gel contain additional 53 bp of flanking primer sequence).

nine cases in addition to normal mdm2, and in two cases, which were lacking the full length mdm2, as the only mdm2 cDNA.

The mdm2-b splice variant of 654 bp was found in one case of the normal ovarian tissues. The most frequent splice variant was the 221 bp mdm2 cDNA, which was seen in 40% of normal ovarian tissues. The 259 bp splice variant was found in four cases (20%).

Interestingly, seven out of 12 patients who had undergone surgery for a benign disease, mostly uterine fibroids, had expression of only normal mdm2. In contrast, none of the six patients who had undergone surgery for a gynecological cancer had exclusively expression of the full-length mdm2 in the normal ovarian tissue. In five of these cases we found one or several splice variants, besides the full-length mdm2 expression, and in one case only two splice variants and no full length mdm2 cDNA. This result reached statistical significance. The finding of mdm2 RNA splicing in the unaffected normal ovary was significantly correlated with the diagnosis of a gynecological cancer, either in the contralateral ovary, or in the uterine cervix or endometrium (p=0.017).


115

Table 21: mdm2 alternative and aberrant RNA splicing in normal ovarian tissue

Case
ID #

Age of
patient

Diagnosis at
surgery / autopsy

Meno-pausal
status

Day of
menstrual cycle

full length
mdm2

mdm2-b
654 bp

mdm2
221 bp

Other
splice variants

 

 

 

 

 

 

 

 

 

ov51

41

endometrial cancer

pre

12

+

-

+

365

ov56

48

unknown

?

?

+

-

-

-

ov63

54

endometrial cancer

post

-

+

-

+

279, 259

ov65

37

cervical cancer

pre

42

-

-

-

279, 259

ov97

39

unknown

pre

19

+

-

-

-

ov98

37

uterine fibroids

pre

7

+

-

-

-

ov105

40

uterine fibroids

pre

20

+

-

-

-

ov108

39

uterine fibroids

pre

10

+

-

+

-

ov119

61

uterine fibroids

post

-

+

-

-

-

ov120

37

uterine fibroids

pre

2nd half

+

-

-

-

ov127

39

uterine fibroids

pre

23

+

-

-

-

ov172

41

uterine fibroids

pre

11

+

-

-

-

ov214

39

uterine fibroids

pre

?

+

-

+

-

402A

50

ovarian cancer Ia

?

?

+

-

+

-

2150

60

normal ovary

post

-

+

-

+

259

3106

43

cervical cancer

pre

?

+

-

+

52

3119

41

liver failure

pre

?

-

+

-

-

3294

42

normal ovary

pre

?

+

-

-

442

3539

-

contralateral ovarian cancer Ia

?

?

+

-

+

-

73595

50

uterine fibroids

post

-

+

-

-

-

 

 

 

 

 

 

 

 

 

n
(%)

-

-

-

-

18
(90)

1
(5)

8
(40)

6
(30)

4.2.7 In vitro expression of p53 and MDM2 splice variant proteins

Transient transfection of HeLa cells with the pcDNA3 expression vector containing the mdm2 cDNA of interest was used for expression of MDM2 and p53 proteins (Fig. 29). The cDNA sequence had been inserted into the pcDNA3 plasmid such that it came under the control of the T7 RNA polymerase promoter (pT7). The HeLa cells were previously infected with the recombinant vaccinia virus vTF7-3 which expresses the bacteriophage T7 RNA polymerase gene. During incubation, the


116

gene of interest is transcribed with high efficiency by T7 RNA polymerase (Panicali and Paoletti, 1982, Mackett et al. 1982, Moss and Flexner, 1987).

A full length p53 cDNA, a full length mdm2 cDNA, and three splice variants of 654 bp, 351 bp, and 52 bp were amplified by PCR using a primer pair designed for cloning the cDNA of interest into a pcDNA3 expression-vector. The pcDNA3 expression vector had been modified such that it contained a myc-epitope respectively a HA-epitope at the 3‘ end of the DNA-insert. The amplified DNA was ligated into the pcDNA3 vector and transformation of electro-competent E. coli bacteria was carried out through electroporation. The correct size of the insert was verified by digestion of the plasmid by the NcoI restriction enzyme. (Fig. 8b).

HeLa cells were infected with the vaccinia virus strain vTF7-3 and subsequently, transient transfection of the infected HeLa-cells with the expression vector plasmid was carried out through incubation with a DNA-liposome-complex (Fig. 9).

After harvesting the cells, proteins were analyzed by SDS-polyacrylamide gel electrophoresis. We were able to express the full length p53 protein (53 kDa), the full length MDM2 protein (90 kDa), and the protein of the mdm2-b-splice variant (40kDa) in vitro in HeLa-cells (Fig. 29). The clones containing the splice variants 351 bp and 52 bp did either not express a protein, or it was undetectable due to small amounts or rapid degradation.

Co-immunoprecipitation was carried out by transient transfection of vTF7-3 infected HeLa-cells with both the full length p53 cDNA and the mdm2-b splice variant cDNA, respectively with p53 and full length mdm2 cDNA as a control, followed by SDS-PAGE and immunoprecipitation with the anti-myc respectively anti-HA-antibody. We were able to demonstrate that the full length MDM2 protein binds to the p53 protein (results not shown). However, co-expression and co-immuno-precipitation of the p53 protein and the mdm2-b splice variant protein were not successful so far. Further experiments will be needed to clarify, whether p53 binds to the mdm2 splice variant proteins.

4.2.8 Correlation of mdm2 and p53 alterations

Out of 92 ovarian cancer cases analyzed for mdm2 splice variants, 50 (54%)
had p53 mutations, and 52 (57%) had p53 overexpression including 13 cases with
wildtype p53 sequence. No correlation was found between p53 mutation status and


117

Fig. 29: Cytoplasmic expression of p53 and MDM2 proteins by transient transfection of pcDNA3 vector constructs into vaccinia virus infected HeLa cells
A:
Expression of the full length p53 protein (53 kDa). B: Expression of the full length MDM2 protein (90 kDa).
C: Expression of a 40 kDa protein of the splice variant mdm2-b (654 bp). No protein could be expressed from the mdm2 splice variants of 351 bp and 52 bp. Proteins were analyzed by Western blotting and anti-myc respectively anti-HA immunodetection.


118

mdm2 splicing. In cases with p53 mutation, 22% (11/50) had no detectable full length mdm2, 72% (36/50) had splice variants, 48% (24/50) expressed the mdm2-b splice variant and 12% (6/50) expressed the 221 bp splice variant.

Out of 52 cases with p53 protein overexpression, 7 (14%) had no detectable full length mdm2, 41 (79%) had splice variants, and 9 (17%) expressed the 221 bp mdm2 variant. The only significant difference was seen for the mdm2-b splice variant which was present in 52% of the cases with p53 overexpression but only in 28% of cases with no immunohistochemically detectable p53 protein (p=0.018).

In cases with p53 overexpression despite wildtype sequence, mdm2 splicing was seen more frequently (11/13, 85%) than in cases with normal p53 expression and wildtype p53 sequence (19/29, 65%). However, the results did not reach statistical significance.

Table 22: p53 alterations in comparison to mdm2 RNA splicing

p53 sequence /
p53 protein

no mdm2
expression
n (%)

full length
mdm2
n (%)

full length and
splice variants
n (%)

only splice
variants
n (%)

Total

n (%)

wildtype/
normal expression

3 (10)

7 (24)

18 (62)

1 (3)

29 (32)

wildtype/
overexpression

0 (0)

2 (15)

10 (77)

1 (8)

13 (14)

mutation/
normal expression

2 (18)

3 (27)

3 (27)

3 (27)

11 (12)

mutation/
overexpression

3 (8)

6 (15)

27 (69)

3 (8)

39 (42)

 

 

 

 

 

 

 

8 (9)

18 (20)

58 (63)

8 (9)

92 (100)
p = 0.23

4.2.9 Correlation of mdm2 alterations with histopathological and clinical data

If mdm2 results were classified into full length mdm2, altered, and no expression of mdm2, we did not find any correlation with clinical or histological characteristics of the ovarian tumors (Table 23). A comparison of the most frequent splice variants though, the 654 bp and the 221 bp variant, showed some interesting correlations.


119

Table 23: Clinico-pathologic characteristics of patients with ovarian cancer and mdm2 alterations

 

 

 

 

 

 

 

mdm2 Alterations

Splice variant 654 bp

Splice variant 221 bp

 

 

 

 

 

 

(mdm2-b)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Characteristic

mdm2

altered

no mdm2

P

absent

present

P

absent

present

P

 

 

full length

mdm2

expression

value*

 

 

value*

 

 

value*

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Patients - no. (%)

18 (20%)

66 (71%)

8 (9%)

 

54 (59%)

38 (41%)

 

77 (84%)

15 (16%)

 

 

 

 

 

 

 

 

 

 

 

 

 

Median age - yrs

58

55

50

 

55

57

 

55

57

 

 

 

 

 

 

 

 

 

 

 

 

 

Histological Subtype - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Serous

12 (67%)

25 (39%)

3 (38%)

0.10

24 (44%)

16 (43%)

0.90

35 (46%)

5 (36%)

0.50

 

Other

6 (34%)

40 (61%)

5 (62%)

 

30 (56%)

21 (57%)

 

42 (54%)

9 (46%)

 

 

 

 

 

 

 

 

 

 

 

 

 

Clinical stage - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FIGO I

6 (33%)

19 (29%)

3 (38%)

0.50

20 (37%)

8 (21%)

0.066

20 (26%)

8 (53%)

0.017

 

FIGO II

2 (11%)

3 (5%)

0

 

4 (7%)

1 (3%)

 

4 (5%)

1 (6%)

 

 

FIGO III

8 (44%)

36 (54%)

3 (38%)

 

24 (44%)

23 (61%)

 

41 (53%)

6 (40%)

 

 

FIGO IV

2 (11%)

8 (12%)

2 (25%)

 

6 (11%)

6 (16%)

 

12 (16%)

0

 

 

 

 

 

 

 

 

 

 

 

 

 

Grade of Differentiation - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

(well)

7 (39%)

11 (17%)

3 (37%)

0.41

17 (31%)

4 (10%)

0.04

17 (22%)

4 (27%)

0.31

 

II

(moderate)

3 (17%)

23 (35%)

0

 

14 (26%)

12 (32%)

 

20 (26%)

6 (40%)

 

 

III

(poor)

8 (44%)

32 (48%)

5 (63%)

 

23 (43%)

22 (58%)

 

40 (52%)

5 (33%)

 

 

 

 

 

 

 

 

 

 

 

 

 

Residual Tumor - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No residual tumor

13 (76%)

36 (58%)

5 (63%)

0.65

37 (76%)

17 (45%)

0.004

44 (59%)

10 (77%)

0.12

 

< 2 cm

2 (12%)

16 (26%)

2 (25%)

 

8 (16%)

12 (32%)

 

17 (23%)

3 (23%)

 

 

> 2 cm

2 (12%)

10 (16%)

1 (13%)

 

4 (8%)

9 (24%)

 

13 (18%)

0

 

 

 

 

 

 

 

 

 

 

 

 

 

Lymph nodes - no. (%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Negative

6 (35%)

13 (21%)

1 (13%)

0.89

14 (29%)

6 (16%)

0.45

17 (23%)

3 (23%)

0.99

 

Positive

4 (24%)

20 (32%)

3 (37%)

 

14 (29%)

13 (34%)

 

23 (31%)

4 (31%)

 

 

Not resected

7 (41%)

29 (47%)

4 (50%)

 

21 (42%)

19 (50%)

 

34 (46%)

6 (46%)

 

 

 

 

 

 

 

 

 

 

 

 

 

DNA-ploidy - no. (%) (n=88)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Diploid

8 (44%)

33 (55%)

4 (57%)

0.51

26 (51%)

19 (51%)

0.97

36 (49%)

9 (60%)

0.45

 

Aneuploid

10 (56%)

30 (45%)

3 (43%)

 

25 (49%)

18 (49%)

 

37 (51%)

6 (40%)

 

 

 

 

 

 

 

 

 

 

 

 

 

S-Phase-Fraction - no. (%) (n=88)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

low ( < 5%)

5 (28%)

17 (27%)

1 (14%)

0.72

17 (33%)

6 (16%)

0.016

16 (22%)

7 (47%)

0.056

 

intermediate ( 5% - 14.4%)

8 (44%)

27 (43%)

5 (71%)

 

24 (47%)

16 (43%

 

32 (44%)

8 (53%)

 

 

high ( > 14.5%)

5 (28%)

19 (30%)

1 (14%)

 

10 (20%)

15 (41%)

 

25 (34%)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Treatment - no. (%) **

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cisplatin - Cyclophosphamide

3 (17%)

14 (21%)

2 (25%)

1.0

9 (17%)

10 (26%)

0.19

16 (21%)

3 (20%)

0.66

 

Carboplatin - Cyclophosphamide

4 (22%)

13 (13%)

1 (12%)

 

9 (17%)

9 (24%)

 

16 (21%)

2 (13%)

 

 

Other regimens

4 (22%)

12 (18%)

1 (12%)

 

12 (22%)

5 (13%)

 

14 (18%)

3 (20%)

 

 

No chemotherapy

5 (28%)

21 (32%)

3 (38%)

 

16 (30%)

13 (34%)

 

25 (33%)

4 (27%)

 

 

No information - no.

2 (11%)

6 (9%)

1 (12%)

 

8 (15%)

1 (3%)

 

6 (8%)

3 (20%)

 

 

 

 

 

 

 

 

 

 

 

 

 

* Chi-Square test for two categories, Mantel-Haenszel test for three or more categories
** Chemotherapy regimens are described in the methods section


120

Presence of the 654 bp splice variant (mdm2-b) was significantly correlated with poor grade of differentiation (p=0.04), a residual tumor mass after surgery (p=0.004) and high S-phase fraction (p=0.016). It was also more frequent in tumors of advanced stage (p=0.066) (Table 23). The mdm2-b splice variant thus seems to be significantly associated with histologically and clinically more aggressive tumors.

As opposed to this the presence of the splice variant 221 bp was significantly correlated with early stage tumors (p=0.017), and was more frequently identified in grade I or II tumors, tumors without residual mass after surgery, and tumors with low S-phase fraction, though the latter results did not reach statistical significance (Table 23) (Table 23).

4.2.10 mdm2 alterations and clinical outcome in ovarian cancer

mdm2 splice variant findings were furthermore compared with the response to a platinum based chemotherapy (Table 24). Details of chemotherapy regimen and evaluation of response were described in the methods section. If full length mdm2 was present in the ovarian carcinomas, the majority of tumors (64%) were sensitive to

Table 24: Response to platinum-based chemotherapy in ovarian cancer depending on mdm2 alternative splicing

 

Chemotherapy-Response

 

 

 

 

mdm2 transcript

Sensitive
n (%)

Resistant
n (%)

Refractory
n (%)

Total
n (%)

p-value

 

 

 

 

 

 

mdm2 full length present

27 (64)

3 (7)

12 (29)

42 (84)

 

mdm2 full length absent

2 (25)

5 (63)

1 (13)

8 (16)

 

 

 

 

 

 

 

 

29

8

13

50

p = 0.001

 

 

 

 

 

 

no mdm2 expression

1 (25)

2 (50)

1 (25)

4 (8)

 

normal mdm2 expression

8 (80)

0 (0)

2 (20)

10 (20)

 

normal and splice variants

19 (59)

3 (9)

10 (31)

32 (64)

 

only splice variants

1 (25)

3 (75)

0 (0)

4 (8)

 

 

 

 

 

 

 

 

29 (58)

8 (16)

13 (26)

50 (100)

p=0.006


121

platinum-based chemotherapy (27/42), whereas in cases with no detectable full length transcript, only 2/8 cases were sensitive (25%), while the majority of cases was resistant or refractory (p=0.001) (Table 24). Similarly, presence of mdm2 splicing in general was significantly correlated with poor response to chemotherapy (p=0.006) (Table 24).

mdm2 alterations were analyzed for their impact on the clinical outcome of the patients by log-rank statistical survival analysis (Fig. 30A-F). Presence or absence of alternative splicing as opposed to full length mdm2 cDNA, absence of mdm2-expression, and the site of RNA splicing at exon-intron boundaries as opposed to cryptic exon splice sites, did not have a significant influence on survival after diagnosis of ovarian cancer (Fig. 30A,B). The size of the splice variants had some influence on survival but did not reach statistical significance (Fig. 30C,D). However, analysis of the splice variant 654 bp showed a trend to poor clinical outcome for patients with expression of this splice variant in the ovarian tumor (p=0.17, Fig. 30E). The only significant difference was seen for patients with expression of the 221 bp splice variant (Fig. 30F), who had a better survival prognosis than patients without expression of the 221 bp splice variant (p=0.048).


122

Fig. 30: A-B: Clinical outcome of ovarian cancer patients dependent on mdm2
alternative RNA splicing (n=92) as analyzed by RT-PCR and DNA sequencing.
A)
Overall survival in patients with expression of the full length mdm2 RNA transcript,
versus mdm2 splice variants, versus absence of mdm2 RNA expression.
B) Overall survival in patients with aberrant splicing in the donor and acceptor or only
the acceptor site, versus patients with full length mdm2 and/or alternative splicing at
exon/intron boundary of exon 3 and 9 (mdm2-a, 888 bp) or exon 3 and 11 (mdm2-b, 654 bp).


123

Fig.30 C-D: Clinical outcome of ovarian cancer patients dependent on mdm2 RNA alternative splicing (n=92) as analyzed by RT-PCR and DNA sequencing.
C)
Overall survival in patients with expression of small splice variants of <300 bp versus expression of the mdm2-b splice variant (654 bp) or full length mdm2
D) Overall survival in patients with expression of small splice variants (<300 bp) versus expression of mdm2 splice variants of >300 bp in the presence or absence of the full length mdm2 transcript. Cases with expression of the the mdm2-b splice variant (654 bp) were not included in this analysis.


124

Fig.30 E-F: Clinical outcome of ovarian cancer patients dependent on mdm2 RNA alternative splicing (n=92) as analyzed by RT-PCR and DNA sequencing.
E)
Overall survival in patients with expression of the mdm2-b splice variant (654 bp) versus absence of the mdm2-b splice variant.
F) Overall survival in patients with expression of a small splice variant of 221 bp was correlated with early stage (FIGO I/II) and marginally significant with a better clinical outcome.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.

DiML DTD Version 2.0
Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML - Version erstellt am:
Thu Aug 15 15:17:40 2002