patterns of p53 gene mutations in head and neck cancer: full-the levels of p53 protein using ihc, on...

Patterns of p53 Gene Mutations in Head and Neck Cancer: Full-Length Gene Sequencing and Results of Primary Radiotherapy1

M. Elizabeth Saunders, Robert MacKenzie,Rob Shipman, Edmee Fransen, Ralph Gilbert,and Richard C. K. Jordan2

Faculty of Dentistry, University of Toronto, Toronto [M. E. S.,R. C. K. J.]; Visible Genetics Inc., Toronto, Ontario [M. E. S., R. S.];Toronto-Sunnybrook Regional Cancer Centre, Toronto, Ontario[R. M., E. F., R. G., R. C. K. J.]; and Departments of Dentistry[R. C. K. J.] and Laboratory Medicine [R. C. K. J.], Sunnybrook andWomen’s College Health Sciences Centre, Toronto, Ontario M4N3M5, Canada

ABSTRACTp53 gene alterations are common in head and neck

cancers, but their prognostic value has not been clearlyestablished. Despite evidence in other cancers that sequenc-ing of the entire p53 coding region provides prognosticinformation, full-length p53gene sequencing has rarely beenperformed in head and neck cancers. In this study, p53 wasassessed in a series of 42 pretreatment biopsies from patientswith laryngeal carcinomas by full-length gene sequencingand by immunohistochemistry (IHC). Associations amongp53genotype, protein expression, and local recurrence wereassessed in 35 irradiated patients followed for a minimum of5 years. DNA was extracted from formalin-fixed, paraffin-embedded biopsies, and exons 2–11 of thep53 gene wereindividually amplified by PCR and then directly sequenced.IHC was performed to detect mutant and wild-type p53protein using the DO7 monoclonal antibody. p21 proteinexpression was assessed using the EA1 monoclonal antibody.Twenty genetic alterations were observed in 42 tumors(48%). Four of these alterations (20%) occurred outsideexons 5–8. There was a significant association betweenp53gene and protein status (x2 5 4.18,P 5 0.04), although thecorrelation was weak (f coefficient 5 20.327). Althoughlocal relapse following radiation was significantly associatedwith nodal status, no correlations were observed betweenp53 status (gene or IHC) and local recurrence followingradiation therapy, based on the Kaplan-Meier method.These results show that p53 mutations are common in la-ryngeal carcinomas and that a proportion occur outside

traditionally examined regions. The lack of correlation be-tween p53 status and local control suggests that this markeris not as powerful as traditional prognostic factors, such aslymph node status.

INTRODUCTIONHNSCC3 is the sixth most common cancer in the world,

accounting for 4% of cancers in men and 2% of cancers inwomen (1). Worldwide,;500,000 new cases of HNSCC arediagnosed annually (2). The current management for HNSCCuses radiation therapy and surgery, either alone or in combina-tion (3). Despite treatment advances that have resulted in reduc-tions in patient morbidity, overall 5-year survival rates forHNSCC have remained stubbornly low, at;30–40%, for thepast several decades (4, 5). These results are clearly disappoint-ing and support the need for further studies on the biology ofthis disease.

Alteration of thep53 gene is the single most commonlyreported genetic abnormality in a number of cancers, includingthose arising from the head and neck (6, 7). Normally, p53encodes for a nuclear phosphoprotein, which acts as a sequence-specific transcription factor and is involved in cell cycle regu-lation and cellular response to DNA damage (8). In this setting,p53 protein levels rise, leading to either cell cycle arrest orapoptosis (9, 10). p53-dependent G1-S arrest is mediated, inpart, by transactivation of thep21waf1/cip1gene, which, in turn,promotes cell cycle arrest by binding to and inhibiting cyclin-dependent kinases (11).

In tumors, the prevalence of p53 alterations varies, rangingfrom 10 to 60% in most cancers and topping 80% in somesubsets (7). In HNSCC the reported frequency of p53 alterationsranges from 20 to 90%, depending on the methodologies toassess p53, type of tumor material, and heterogeneity of tumorsites examined. In most studies, investigators have determinedthe levels of p53 protein using IHC, on the basis that mutantprotein is stable but that the half-life of wild-type protein is tooshort to permit detection (12, 13). Although IHC can provideimportant information about protein expression, some studieshave shown an inconsistent relationship between p53 proteinlevels and gene status (14, 15). A number of explanations mayaccount for these discrepancies, including differences in antigenretrieval techniques as they are applied to routinely processedtissue specimens, the effects of non-missense mutations onprotein expression, and alternate mechanisms of p53 proteinstabilization (16–18). Fewer studies have examinedp53 at thegene level, in part because it is more laborious, technically moredemanding, and frequently not applicable to routinely processed

Received 1/22/99; revised 4/21/99; accepted 6/1/99.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisementin accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 Supported by Sunnybrook Research Trust (Medical Research Councilof Canada) Grant UI-14291 and by Visible Genetics Inc.2 To whom requests for reprints should be addressed, at H-126, Sunny-brook and Women’s College Health Sciences Centre, 2075 BayviewAvenue, Toronto, Ontario M4N 3M5 Canada. Phone: (416) 480-4436;Fax: (416) 480-5757.

3 The abbreviations used are: HNSCC, squamous cell carcinoma of thehead and neck; IHC, immunohistochemistry; CT, computed tomograph-ic; TBS, Tris-buffered saline.

2455Vol. 5, 2455–2463, September 1999 Clinical Cancer Research

Research. on January 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

tumor biopsies. Moreover, studies that have examinedp53at thegene level have frequently restricted their analysis to traditional“hot spot” regions in exons 5–8 (19–21). Early studies on anarrow range of tumors showed that the majority of p53 muta-tions occurred within the highly conserved regions in exons 5–8that code for the DNA-binding domain (7). However, it has beenshown that, in some tumor types, up to 25% of gene alterationsmay occur outside exons 5–8, emphasizing the need for exam-ination of all coding exons (17, 22). Moreover, in some cancers,full-length sequencing to identifyp53 gene alteration can pro-vide significant prognostic information (23). Despite this, theprevalence of mutations throughout the entire p53 coding regionhas not been examined in head and neck tumors from a singleanatomical site, such as the larynx.

Mutations in thep53gene may provide cells with a growthadvantage by decoupling them from the normal restrictionsimposed on cellular proliferation in response to DNA damage(24, 25). Consequently, cells with p53 mutations may be moreresistant to therapies that depend on generating DNA damage tokill cancerous cells. The importance of p53 status and responseto radiation therapy in tumor samples is supported by studies oftumors arising in the esophagus, breast, and prostate (23, 26).For example, Prendergastet al. (27) showed that, despite thelow prevalence of p53 alterations in pretreatment tumor speci-mens, p53 mutations and altered protein expression were sig-nificant predictors of local treatment failure in a group of 18node-positive prostate cancers treated by radiation therapy. Re-sults of studies examining the importance of p53 alterations inHNSCC treated by radiation therapy are conflicting; some stud-ies have shown p53 alterations predict local control, but othersshow no relationship (28–32). Differing results may arise fromthe pooling of tumors from sites throughout the head and neck,incomplete analyses, and inconsistent treatment regimes. Theaim of this study was to determine the prevalence ofp53 genealterations in a series of laryngeal carcinomas by direct sequenc-ing of all coding exons. We also examined the relationshipbetweenp53 gene alterations and tumor response to radiationtherapy, as assessed by local control.

MATERIALS AND METHODSCase Selection. Between June 1984 and December 1990,

42 unselected patients with cancer of the larynx were managedby the multidisciplinary head and neck site group at theToronto-Sunnybrook Regional Cancer Center with a policy ofprimary radiotherapy, under which surgery was reserved forsalvage. The diagnostic work-up included examination underanesthesia with biopsy (if not already performed by the referringphysician), complete physical examination, complete bloodcounts, renal and liver function tests, chest X-ray, and laryngealtomograms and/or CT scan of the head and neck. An ultrasoundof the liver was performed in patients with abnormal liverfunction tests, and bone scans were performed in patients withunexplained skeletal pain. Patients were staged according to theUnion International Contre Cancer tumor-node-metastasis clas-sification. There were, in total, 42 pretreatment, formalin-fixed,paraffin-embedded tissue biopsies obtained.

Radiation Therapy. All patients were immobilized in aplastic mask and treated with60Co on a 6-MV linear accelerator.

The primary tumor and bilateral neck nodes were irradiatedthrough lateral opposed photon beams. The supraclavicularnodes were treated with a matched anterior photon field. Theinitial treatment volumes, including primary tumor, involvednodes, and potential sites of microscopic spread, were treated toa dose of 4600 cGy in 23 fractions over 4.5 weeks to the larynxand regional nodes. Sites of gross residual disease that were#3cm in maximum diameter were boosted with an additional 2000cGy in 10 fractions over 2 weeks, whereas those that were.3cm were boosted with an additional 2400 cGy in 12 fractionsover 2.5 weeks. All patients, therefore, received a total dose ofeither 6600 or 7000 cGy.

Monitoring and Follow-Up. Patients were assessedclinically at monthly intervals. A CT scan of the head and neckwas obtained 8–12 weeks posttreatment. Complete responderswere followed every 3 months for the first 2 years, every 6months for the next 3 years, and annually thereafter. ChestX-rays were performed at 3 and 12 months and annually there-after. Suspected sites of local recurrence were evaluated withCT scan and confirmed by aspiration cytology or by biopsy.Disease identified on metastatic survey was not biopsied.

Radiation Therapy Study Cohort. Correlations be-tween p53 status (gene and protein) and treatment outcomeswith radiotherapy were limited to a subset of 35 patients whowere followed for a minimum follow-up of 5 years or until theyexperienced an in-treatment field recurrence. This cohort wascharacterized by a median age of 62 years (range, 45–76 years)and a male:female ratio of 4:1. Distribution by tumor-node-metastasis stage is summarized in Table 1.

Genomic DNA Preparation from Paraffin Sections.Three 5-mm sections of each sample were cut with a microtomeand placed on glass slides. Each slide was heated to 65°C to meltthe paraffin, and then the tissue was removed with a scalpel andplaced in 12ml of 103 lysis buffer [100 mM Tris-HCl (pH 8.0),500 mM KCl, 25 mM MgCl2, and 4.5% Tween 20], 98ml ofdistilled H2O, and 10ml of 20 mg/ml proteinase K (Sigma).Following overnight incubation at 55°C, the proteinase K wasinactivated by heating the samples for 5 min at 94°C. PCR wasthen performed on the solubilized DNA recovered in the super-natant.

PCR Amplification. Exons 2–11 of thep53 gene wereamplified separately using p53 Gene Analysis Kits (VisibleGenetics Inc., Toronto, Canada). Briefly, 3ml of genomic DNAwere added to a 25-ml PCR mixture according to manufacturer’sinstructions and then cycled through 35 cycles consisting of94°C for 30 s, 60°C for 30 s, and 72°C for 60 s on a thermo-

Table 1 Distribution by site and stage of 35 patients treated withprimary radiotherapy

T stage

Glottis Supraglottis

N0 N1 N0 N1

T1 16 0 2 0T2 5 0 1 1T3 1 0 0 3T4 1 0 4 1

Subtotal 23 0 7 5

2456p53 Gene Mutations in Head and Neck Cancer



cycler (MJ Research). PCR products were then visualized byelectrophoresis of 2ml of reaction product in a 1% agarose gel.

Cycle Sequencing. PCR products were sequenced usingnested CY5.5-labeled primers provided in the p53 Gene Anal-ysis Kit (Visible Genetics Inc.). Two to 8ml of each PCRproduct were used in a sequencing reaction according to man-ufacturer’s instructions, and the products were analyzed usingthe MicroGene Blaster Automated DNA electrophoresis unit(Visible Genetics Inc.). All samples were initially screened bysequencing in either the 59 or 39 directions. All gene alterationswere confirmed by repeat sequencing in the opposite directionand then reconfirmed by further reamplification and direct se-quencing steps.

IHC. Briefly, 5-mm serial sections were cut and mountedon glass slides coated with 2% aminopropyltrioxysilane (SigmaChemical Co., St. Louis, MO) in acetone. Sections were dew-axed in xylene and rehydrated in graded ethanols. Endogenous

peroxidase activity was blocked by immersion in 0.3% metha-nolic peroxide for 15 min. Immunoreactivity of the target anti-gens was enhanced using pressurized heat antigen retrieval(pressure cooking; Ref. 33). The sections were placed in apressure cooker containing 0.01M sodium citrate buffer (pH6.0), heated to 130°C for 2 min, and then cooled. p53 proteinexpression was determined by incubating the tissue sectionswith the monoclonal antibody DO7 (Novocastra, Newcastle,United Kingdom) diluted 1:500 in 13 TBS. This antibodyrecognizes both wild-type and mutant forms of the p53 protein.For analysis of p21 protein expression, tissue sections wereincubated with EA1 monoclonal antibody (Oncogene Sciences,Cambridge, MA) diluted 1:200 in TBS. All incubations werecarried out overnight at 4°C. The sections were then incubatedwith a biotinylated secondary antimouse antibody diluted 1:200in TBS for 60 min, followed by the application of preformedavidin-biotin complex (Dakopatts, Denmark) for 60 min. The

Table 2 Summary ofp53 gene mutations and results of IHC for p53 and p21 proteinsa

Patient reference no. p53 mutation Codon Exon Amino acid change p53 IHCb p21 IHCb % tumor in biopsy

12 16-bp del 85–90 4 Frameshift 0 1 9037 4-bp ins 86–87 4 Frameshift 0 111 804 G3T 135 5 Cys3Phe 0 0 958 G3A 173 5 Val3Met 111 1 80

20 G3T 176 5 Cys3Phe 11 11 7518 A3T 179 5 His3Leu 111 1 8027 2-bp del 189 6 Frameshift ND ND 701 G3A 216 6 Val3Met 111 11 80

24 A3G 220 6 Tyr3Cys 111 1 8040 A3G 220 6 Tyr3Cys 111 111 1030 G3A 248 7 Arg3Gln 111 11 6011 C3T 248 7 Arg3Trp 111 0 7031 G3T 249 7 Arg3Ser 111 111 6014 G3C 273 8 Arg3Pro 111 1 4019 G3C 274 8 Val3Leu 111 11 2029 G3T 274 8 Val3Phe ND ND 4036 C3A 278 8 Pro3His 111 111 7032 G3A 286 8 Glu3Lys 111 111 8038 25-bp del 315–322 9 Frameshift 1 11 502 G3A 342 10 Arg3Gln 111 111 803 wt 0 1 705 wt 0 111 806 wt 0 111 607 wt 0 111 409 wt 1 11 20

10 wt 111 1 6013 wt 0 1 8015 wt 11 1 8016 wt 1 0 6017 wt 111 11 6021 wt 0 11 6522 wt 0 1 7023 wt ND ND 5025 wt 0 111 9026 wt 11 1 8028 wt 111 1 8033 wt 111 1 9034 wt 111 1 4035 wt 0 0 6039 wt 1 11 4041 wt 111 111 8042 wt 0 11 60

a del, deletion; ins, insertion; ND, not done; wt, wild type.b Immunohistochemical staining categories: 0, 5–9% tumor cells stained;1, 10–24%;11, 25–49%;111, $50%.

2457Clinical Cancer Research



bound complexes were visualized by the application of a 0.05%solution of 3,39-diaminobenzidine (pH 7.6; Sigma) containing0.3% hydrogen peroxide as a substrate. Following incubation,the sections were washed and then lightly counterstained withhematoxylin, dehydrated, and coverslipped.

Quantification. Scoring of p53 and p21 immunostainingwas performed by counting the number of cells showing nuclearstaining as a proportion of the total tumor cell population,expressed categorically as follows: 0, 0–9% tumor cells stained;1, 10–24% tumor cells stained;11, 25–49% tumor cellsstained; and111, $50% tumor cells stained. For p53, tumorsshowing.10% positive cells were considered to show abnor-mal protein staining.

Statistical Analysis. Associations between p53 gene andprotein status and with p21 protein expression were assessed byx2 analysis. The strength of association was assessed by thefcorrelation coefficient. For the radiation therapy cohort, Kaplan-Meier method was used to model time to tumor recurrence as afunction of p53 mutational status and p53 protein expression.

RESULTSWe identified p53 gene alterations in 20 of 42 (48%)

biopsies of laryngeal carcinoma. These results are summarizedin Tables 2 and 3. These alterations were primarily point mu-tations resulting in nonconservative amino acid substitutions(Fig. 1a). In addition, there were three insertions/deletions butno nonsense or splice site alterations (Fig. 1b). Genetic alter-ations were detected throughout the gene, from codon 85 to 355.Of the 20 alterations, four (20%) were outside exons 5–8. Thepattern of gene mutations outside these exons differed from thepattern inside. In contrast to the changes inside exons 5–8,which were mostly missense changes, the gene alterations out-side exons 5–8 were mostly insertions and deletions. The pro-portion of tumorversusnormal tissue in each biopsy rangedfrom 10 to 90% (Table 2) with p53 mutations detected in tumorsthroughout this range.

Using IHC, p53 protein overexpression (Fig. 2) was ob-served in 26 of 39 samples (67%; Table 4). IHC was notperformed on three cases due to tissue depletion. p53 mutationalstatus was significantly associated with protein expression (x2

5 4.18, P 5 0.04), but this correlation was weak (f coeffi-cient 5 20.327). Three of 18 (17%) samples withp53 genealterations did not show p53 protein overexpression. In contrast,

11 of 21 (52%) of samples without detectablep53 gene alter-ations showed overexpression of the protein.

p21 staining was seen in the nuclei of tumor cells (Fig. 3).There was a heterogeneous distribution of p21 staining in tumorcells, with the proportion of positive cells ranging from 5 to 75%in the tumors. There were no correlations between the levels ofp21 protein and p53 protein expression (x2 5 3.457,P 5 0.33)or p53 gene status (x2 5 1.627,P 5 0.65).

The radiation therapy study cohort of 35 patients treatedwith radical radiation therapy was followed for a median of 4years. The 5-year cause-specific, disease-free, and overall sur-vival rates were 91, 77, and 63%, respectively. Local recurrencewas documented in 8 patients (Table 5). A univariate analysis ofprognostic factors identified T and N stages as significant pre-dictors of local failure. In this analysis, neither thep53 genestatus (Fig. 4) nor the combination ofp53 gene and IHC status(Fig. 5) emerged as a significant predictor of local failure.

DISCUSSIONp53 gene alterations are the single most common genetic

abnormality in all cancers, including those arising in the headand neck. The reported prevalence of p53 mutations in head andneck cancers varies from 20 to 91% (34–37), and this range mayreflect the heterogeneity of tumors that arise in this region (38).Our study examined a cohort of tumors from a well-definedanatomical site with the head and neck, the larynx, and foundgene alterations in 48% of patients. These data support theconcept that alterations of thep53 gene play an important rolein the development of a large proportion of tumors at this site(19, 39).

Fig. 1 p53 sequence trace from tumor samples. Comparisons betweenthe wild-type (normal) and mutant (tumor) sequences are listed beloweach trace.a, exon 10 sequence traces from patient 2 (sample 32)showing a G3A transition at codon 342, resulting in an arginine toglutamine amino acid substitution.b, exon 4 sequence trace from patient12 (sample 32) showing a 16-bp deletion at codons 85–90, resulting inpremature protein truncation.

Table 3 Summary ofp53 gene alterations by exon and predictedamino acid changea

ExonNo. of

mutations Type Amino acid changes

4 2 13 4-bp ins Frameshift1 3 16-bp del Frameshift

5 4 43 missense Nonconserved6 4 3 missense Nonconserved

1 3 2-bp del Frameshift7 3 43 missense 3 nonconserved8 5 53 missense 3 nonconserved; 2 conserved9 1 13 25-bp del Frameshift

10 1 13 missense Nonconserveda del, deletion; ins, insertion.




Initial studies of thep53gene showed that most mutationswere clustered within the conserved regions encoding the DNA-binding domain (8). We found that 20% of allp53 gene muta-tions occur outside exons 5–8. This is similar to the frequencyof 33% in these regions that was found in the only other studyto examine HNSCC forp53 gene mutations outside the DNA-binding domain (37). Moreover, these results are similar tostudies in breast (17, 23, 40), hepatocellular (41), ovarian, andnon-small cell lung (22) carcinomas showing that 10–25% ofp53 gene alterations occur outside traditionally examined hotspot regions.

Similar to tumors at other sites, we found the pattern ofmutations inside and outside exons 5–8 differed. Hartmannetal. (22) examined the promoter region and exons 1–11 of thep53gene in 194 primary breast cancers and found that 18 of 82mutations were outside exons 5–8. Within this region, the genealterations were mainly missense mutations, but outside exons5–8, the majority of changes were insertions, deletions, ornonsense changes. Our results for laryngeal carcinoma are sim-ilar, with the majority of alterations within exon 5–8 missensemutations resulting in nonconserved amino acid substitutions,but outside this region, there was a predominance of insertionsand deletions resulting in protein truncation. It has been sug-

gested that these pattern differences may support the existenceof functional or conformational differences between the centraland peripheral coding regions of thep53 gene (22). Whereasmissense changes in the highly conserved, DNA-binding do-main disrupt protein binding to DNA, missense mutations out-side this region may not significantly alter the protein’s functionin a fashion that gives cells a growth advantage (22). In contrast,insertions, deletions, nonsense mutations, and splice sitechanges may have a more dramatic impact on the overall proteinstructure and conformation that significantly alters the proteinfunction. Our study supports the notion that full-length sequenc-ing of the p53 gene can identify patients with gene mutationsthat would be missed by a more restricted analysis (40).

We found a statistically significant association betweenp53 mutational status and protein overexpression, although thecorrelation was weak (f coefficient5 20.327). We found that17% (3 of 18) of tumor samples that containedp53 genemutations did not show any protein expression by IHC. Two ofthe three false-negative samples (samples 12 and 37) harboredframeshift insertions or deletions that drastically altered aminoacid composition and led to premature protein truncations.These alterations, therefore, may not stabilize the protein suffi-ciently to permit detection by IHC. Alternatively, negative stain-ing of p53 mutant tumor samples may represent inability of theantibody to react with its epitope due to structural alterations ofthe translated protein. The third false-negative sample (sample4) contained a missense mutations that would be predicted toresult in p53 protein stabilization but failed to express p53protein by IHC. One possible explanation for this is that wild-type protein may cause destabilization of mutant p53 protein inits tetrameric form in cells (40).

In contrast, 11 of 21 (52%) of tumors without detectablegene alterations overexpressed p53 protein. There are a number

Fig. 2 IHC to detect tumor cellsoverexpressing p53 protein incase 11. This tumor contained amissense mutation at codon 248in exon 7. (Avidin-biotin com-plex staining,360.)

Table 4 Comparison between results of p53 mutational analysis andp53 protein levels, as assessed by IHCa

p53 mutant p53 wild-type Total

p53 IHCb 1 15 11 26p53 IHC 2 3 10 13Total 18 21 39

a x2 5 4.179;P 5 0.041;f coefficient5 20.327.b 1, .10% tumor cells showing nuclear staining.




of explanations for discrepancy between wild-typep53 genestatus and elevated protein levels. Cells that have stained posi-tive may carry p53 mutations that, due to low copy numbers,may be below the threshold of detection. Alternatively, wild-type p53 protein may be stabilized either by binding to cellularor viral proteins or by disturbances in the ubiquination pathwaythat would otherwise degrade p53 (8). The functional signifi-cance of overexpressed wild-type p53 protein in the develop-ment of laryngeal carcinomas is not known because IHC cannotdifferentiate functionally active from inactive protein (13). Fi-nally, technical issues related to the use of routinely processedbiopsy material may also account for a proportion of casesshowing overexpressed protein detected by IHC in the absenceof a p53 gene mutation (31, 42).

p21 plays an important role as a p53-response elementfollowing DNA damage. Furthermore, a number of studieshave now suggested that p21 is important in cell differenti-ation and senescence pathways by showing that the topo-graphical distribution in a number of tissues, includingesophageal epithelia, is consistent with proliferation arrestand differentiation (43). We found no correlation betweenp53 gene status or protein levels and p21 protein levels. Thisis similar to other studies, including those examining laryn-geal carcinomas that also found an inconsistent relationshipbetween p21 protein levels and p53 status (44). The relation-ship between p21 and p53 in cancers, however, may be tissueand tumor type specific. In low-grade gliomas, p21 proteinlevels are independent of p53 status, but this is not so forglioblastoma multiforme (45). Other tissues in which p21levels are independent of p53 include ovarian carcinoma (46)and pancreatic carcinoma (47). In contrast, in colorectalcancer, there is a correlation between p21 levels and wild-type p53 (48). We also found that p21 levels were independ-

ent of tumor grade. Two other studies have reported conflict-ing findings regarding p21 and tumor differentiation in headand neck cancers, with one study supporting this association(44) and another finding no relationship (49). In addition, wefound no relationship between p21 levels in laryngeal carci-noma and any clinical parameter or prognostic parameterassociated with the disease. Our results differ from those ofErber et al. (49), who found that overexpression of p21protein was associated with increased disease recurrence,shorter disease-free survival, and shortened overall survivaland may reflect the smaller number of treatment failures inour series.

Despite evidence in other cancers that p53 alterations ateither the genetic or the protein levels are predictive oftreatment failure following radiation therapy, their predictivevalue in HNSCC has not been established (23, 26). In thisstudy, no correlation was identified between p53 status at thegene or protein level and local recurrence following radiationtherapy. Similar findings were reported in a study of p53protein levels in 90 laryngeal carcinomas treated by radiationtherapy (31). Similarly, Awwadet al. (50) analyzed 79 headand neck carcinomas treated with radiation and found noassociation between p53 protein overexpression and overalland disease survival. Our results differ from two recentstudies showing a strong correlation between the presence ofp53 gene mutations in cancers from a number of differentsites in the head and neck and response to radiation therapy(28, 32). The lack of correlation in our study may reflect thesmaller number of patients studied in comparison to the twostudies showing significance or may show true tumor sitedifferences. Analysis of a larger cohort of patients withlaryngeal carcinomas is, therefore, needed to validate thisdifference.

Fig. 3 IHC showing nuclearstaining of p21 in tumor cells oflaryngeal carcinoma in case 25.There was no expression of p53protein, and thep53 gene waswild-type. (Avidin-biotin com-plex staining,360.)




In conclusion, we have examined all coding exons of thep53 gene in squamous cell carcinomas from the larynx andfound a high prevalence ofp53gene alterations. We found thata large proportion of gene alterations in these tumors occurredoutside exons 5–8 and that the pattern of gene alterations

differed outside compared with within this region. We foundthat there was weak correlation betweenp53 gene status andprotein expression and that p53 status at the gene or proteinlevel did not correlate with response to radiation therapy, asassessed by local control.

Fig. 4 Kaplan-Meier analysis of local control for 35 patients treatedwith radical radiotherapy according top53 gene status.

Fig. 5 Kaplan-Meier analysis of local control for 35 patients treatedwith radical radiotherapy according to p53 gene and protein status.

Table 5 Summary of p53 status and clinical features for the radiation therapy cohort (total of 35 patients)

Patient reference no. Site Stage Relapsea Gene alteration p53 IHCb p21 IHCc

1 Glottis I None Mutant Positive 112 Supraglottis II None Mutant Positive 1113 Glottis II None Wild-type Negative 14 Glottis 0 None Mutant Negative 05 Glottis III None Wild-type Negative 1116 Glottis I None Wild-type Negative 1117 Supraglottis III Local Wild-type Negative 1118 Supraglottis I None Mutant Positive 19 Glottis I Local Mutant Positive 11

10 Glottis I None Mutant Positive 111 Supraglottis IV None Mutant Positive 012 Glottis I None Mutant Negative 113 Glottis I Local Wild-type Negative 114 Glottis I None Mutant Positive 115 Glottis I None Wild-type Positive 116 Glottis 0 None Wild-type Positive 017 Glottis IV Local Wild-type Positive 1118 Glottis II None Mutant Positive 119 Glottis I None Mutant Positive 1120 Glottis I Local Mutant Positive 1121 Supraglottis IV None Wild-type Negative 022 Glottis I None Wild-type Negative 123 Glottis I None Wild-type ND ND24 Supraglottis IV Local and regional Mutant Positive 125 Supraglottis III None Wild-type Negative 1128 Supraglottis III Local and distant Wild-type Positive 129 Supraglottis IV None Mutant ND ND30 Glottis II None Mutant Positive 1131 Glottis II None Mutant Positive 11132 Glottis II None Mutant Positive 11133 Glottis I None Wild-type Positive 135 Supraglottis I None Wild-type Negative 036 Supraglottis III Local and distant Mutant Positive 11137 Supraglottis IV None Mutant Negative 11138 Glottis I None Mutant Positive 11

a All patients were followed for 5 years after radiation therapy.b Positive,.10% of cells showed p53 protein expression in the nucleus of tumor cells; Negative, no or#10% of tumor cells showed staining

for p53 protein by IHC; ND, not done.c 0, 0–9% tumor cells stained;1, 10–24%;11, 25–49%;111, $50%; ND, not done.




REFERENCES1. Boring, C. C., Squires, T. S., and Tong, T. Cancer statistics, 1991.CA Cancer J. Clin.,41: 19–36, 1991.

2. Parkin, D. M., Pisani, P., and Ferlay, J. Estimates of the worldwideincidence of eighteen major cancers in 1985. Int. J. Cancer,54: 594–606, 1993.

3. Sweeney, P. J., Haraf, D. J., Vokes, E. E., Dougherty, M., andWeichselbaum, R. R. Radiation therapy in head and neck cancer:indications and limitations. Semin. Oncol.,21: 296–303, 1994.4. Borges, A. M., Shrikhande, S. S., and Ganesh, B. Surgical pathologyof squamous carcinoma of the oral cavity: its impact on management.Semin. Surg. Oncol.,5: 310–317, 1989.5. Vokes, E. E., Weichselbaum, R. R., Lippman, S. M., and Hong,W. K. Head and neck cancer. N. Engl. J. Med.,328: 184–194, 1993.6. Nigor, J. M., Baker, S. J., Preisinger, A. C., Jessup, J. M., Hostetter,R., Cleary, K., Bigner, S. H., Davidson, N., Baylin, S., and Devilee, P.Mutations in thep53gene occur in diverse human tumor types. Nature(Lond.), 342: 705–708, 1989.7. Greenblatt, M. S., Bennett, W. P., and Harris, C. C. Mutations in thep53 tumor suppressor gene: clues to cancer etiology and molecularpathogenesis. Cancer Res.,54: 4855–4878, 1994.8. Prokocimer, M., and Rotter, V. Structure and function of p53 innormal cells and their aberrations in cancer cells: projection on thehematologic cell lineages. Blood,84: 2391–2411, 1994.9. Kastan, M. B., Onyekwere, O., Sidransky, D., Vogelstein, B., andCraig, R. W. Participation of p53 protein in the cellular response toDNA damage. Cancer Res.,51: 6304–6311, 1991.10. Shaw, P., Bovey, R., Tardy, S., Sahli, R., Sordat, B., and Costa,J. Induction of apoptosis by wild-type p53 in a human colon tumor-derived cell line. Proc. Natl. Acad. Sci. USA,89: 4495–4499, 1992.11. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., andBeach, D. p21 is a universal inhibitor of cyclin kinases. Nature (Lond.),366: 701–704, 1993.12. Gannon, J. V., Greaves, R., Iggo, R., and Lane, D. P. Activatingmutations in p53 produce a common conformational effect. A mono-clonal antibody specific for the mutant form. EMBO J.,9: 1595–1602,1990.13. Hall, P. A., and Lane, D. P. p53 in tumour pathology: can we trustimmunohistochemistry? Revisited. J. Pathol.,172: 1–4, 1994.14. Calzolari, A., Chiarelli, I., Bianchi, S., Messerini, L., Gallo, O.,Porfirio, and Mattiuz, P. L. Immunohistochemicalvs. molecular biologymethods. Complementary techniques for effective screening of p53alterations in head and neck cancer. Am. J. Clin. Pathol.,107: 7–11,1997.15. Macgeoch, C., Barnes, D. M., Newton, J. A., Mohammed, S.,Hodgson, S. V., Ng, M., Bishop, D. T., and Spurr, N. K. p53 proteindetected by immunohistochemical staining is not always mutant. Dis.Mark., 11: 239–250, 1993.16. Nylander, K., Nilsson, P., Mehle, C., and Roos, G. p53 mutations,protein expression and cell proliferation in squamous cell carcinomasfrom the head and neck. Br. J. Cancer,71: 826–830, 1995.17. Sjogren, S., Inganas, M., Norberg, T., Lindgren, A., Nordgren, H.,Holmberg, L., and Bergh, J. Thep53 gene in breast cancer: prognosticvalue of complementary DNA sequencingversusimmunohistochemis-try. J. Natl. Cancer Inst.,88: 173–182, 1996.18. Ogden, G. R., Kiddie, R. A., Lunny, D. P., and Lane, D. P.Assessment of p53 protein expression in normal, benign, and malignantoral mucosa. J. Pathol.,166: 389–394, 1992.19. Boyle, J. O., Hakim, J., Koch, W., van der Riet, P., Hruban, R. H.,Roa, R. A., Correo, R, Eby, Y. J., Ruppert, J. M., and Sidransky, D. Theincidence of p53 mutations increases with progression of head and neckcancer. Cancer Res.,53: 4477–4480, 1993.20. Brachman, D. G., Graves, D., Vokes, E., Beckett, M., Haraf, D.,Montag, A., Dunphy, E, Mick, R., Yandell, D., and Weichselbaum,R. R. Occurrence ofp53 gene deletions and human papilloma virusinfection in human head and neck cancer. Cancer Res.,52: 4832–4836,1992.

21. Casson, A. G., Mukhopadhyay, T., Cleary, K. R., Ro, J. Y., Levin,B., and Roth, J. A.p53 gene mutations in Barrett’s epithelium andesophageal cancer. Cancer Res.,51: 4495–4499, 1991.

22. Hartmann, A., Blaszyk, H., McGovern, R. M., Schroeder, J. J.,Cunningham, J., De, V. E., Kovach, J. S., and Sommer, S. S.p53genemutations inside and outside exons 5–8: the patterns differ in breast andother cancers. Oncogene,10: 681–688, 1995.

23. Bergh, J., Norberg, T., Sjogren, S., Lindgren, A., and Holmberg, L.Complete sequencing of thep53 gene provides prognostic informationin breast cancer patients, particularly in relation to adjuvant systemictherapy and radiotherapy. Nat. Med.,1: 1029–1034, 1995.

24. Artuso, M., Esteve, A., Bresil, H., Vuillaume, M., and Hall, J. Therole of the ataxia telangiectasia gene in the p53-, WAF1/CIP1(p21)- andGADD45-mediated response to DNA damage produced by ionisingradiation. Oncogene,11: 1427–1435, 1995.

25. Cox, L. S., and Lane, D. P. Tumour suppressors, kinases andclamps: how p53 regulates the cell cycle in response to DNA damage.Bioessays,17: 501–508, 1995.

26. Ribeiro, U. J., Finkelstein, S. D., Safatle-Ribeiro, A. V., Landre-neau, R. J., Clarke, M. R., Bakker, A., Swalsky, P. A., Gooding, W. E.,and Posner, M. C. p53 sequence analysis predicts treatment responseand outcome of patients with esophageal carcinoma. Cancer (Phila.),83:7–18, 1998.

27. Prendergast, N. J., Atkins, M. R., Schatte, E. C., Paulson, D. F., andWalther, P. J. p53 immunohistochemical and genetic alterations areassociated at high incidence with post-irradiated locally persistent pros-tate carcinoma. J. Urol.,155: 1685–1692, 1996.

28. Koch, W. M., Brennan, J. A., Zahurak, M., Goodman, S. N.,Westra, W. H., Schwab, D., Yoo, G. H., Lee, D. J., Forastiere, A. A., andSidransky, D. p53 mutation and locoregional treatment failure in headand neck squamous cell carcinoma. J. Natl. Cancer Inst.,88: 1580–1586, 1996.

29. Wilson, G. D., Richman, P. I., Dische, S., Robinson, B., Daley,F. M., and Ross, D. A. p53 status of head and neck cancer: relation tobiological characteristics and outcome of radiotherapy. Br. J. Cancer,71: 1248–1252, 1995.

30. Sauter, E. R., Ridge, J. A., Litwin, S., and Langer, C. J. Pretreat-ment p53 protein expression correlates with decreased survival in pa-tients with end-stage head and neck cancer. Clin. Cancer Res.,1:1407–1412, 1995.

31. Tan, L. K. S., and Ogden, G. R. p53 overexpression in laryngealcarcinoma is not predictive of response to radiotherapy. Oral Oncol.Eur. J. Cancer,33: 177–181, 1997.

32. Overgaard, J., Sorensen, S. B., Stausbol-Gron, B., Hoyer, M., andAlsner, J. TP53 mutation is an independent prognostic marker for pooroutcome of radiotherapy in squamous cell carcinoma of the head andneck. Int. J. Rad. Oncol. Biol. Phys.,42(1): 21, 1998.

33. Norton, A. J., Jordan, S., and Yeomans, P. Brief, high-temperatureheat denaturation (pressure cooking): a simple and effective method ofantigen retrieval for routinely processed tissues. J. Pathol.,173: 371–379, 1994.34. Nylander, K., Stenling, R., Gustafsson, H., Zackrisson, B., andRoos, G. p53 expression and cell proliferation in squamous cell carci-nomas of the head and neck. Cancer (Phila.),75: 87–93, 1995.35. Brennan, J. A., Boyle, J. O., Koch, W. M., Goodman, S. N., Hruban,R. H., Eby, Y. J., Couch, M. J., Forastiere, A. A., and Sidransky, D.Association between cigarette smoking and mutation of the p53 gene insquamous-cell carcinoma of the head and neck. N. Engl. J. Med.,332:712–717, 1995.36. Mineta, H., Borg, A., Dictor, M., Wahlberg, P., and Wennerberg, J.Correlation between p53 mutation and cyclin D1 amplification in headand neck cancer. Oral. Oncol. Eur. J. Cancer,33: 42–46, 1997.37. Kropveld, A., Rozemuller, E. H., Leppers, F. G., Scheidel, K. C., deWeger, R. A., Koole, R., Hordijk, G. J., Slootweg, P. J., and Tilanus,M. G. Sequencing analysis of RNA and DNA of exons 1 through 11showsp53 gene alterations to be present in almost 100% of head andneck squamous cell cancers. Lab. Invest.,79: 347–353, 1999.




38. Lavieille, J. P., Brambilla, E., Riva-Lavieille, C., Reyt, E., Chara-chon, R., and Brambilla, C. Immunohistochemical detection of p53protein in preneoplastic lesions and squamous cell carcinoma of thehead and neck. Acta Otolaryngol.,115: 334–339, 1995.39. Shin, D. M., Kim, J., Ro, J. Y., Hittelman, J., Roth, J. A., Hong,W. K., and Hittelman, W. N. Activation ofp53 gene expression inpremalignant lesions during head and neck tumorigenesis. Cancer Res.,54: 321–326, 1994.40. Casey, G., Lopez, M. E., Ramos, J. C., Plummer, S. J., Arboleda,M. J., Shaughnessy, M., Karlan, B., and Slamon, D. J. DNA sequenceanalysis of exons 2 through 11 and immunohistochemical staining arerequired to detect all known p53 alterations in human malignancies.Oncogene,13: 1971–1981, 1996.41. Nishida, N., Fukuda, Y., Kokuryu, H., Toguchida, J., Yandell,D. W., Ikenega, M., Imura, H., and Ishizaki, K. Role and mutationalheterogeneity of thep53gene in hepatocellular carcinoma. Cancer Res.,53: 368–372, 1993.42. Dowell, S. P., and Ogden, G. R. The use of antigen retrieval forimmunohistochemical detection of p53 overexpression in malignant andbenign oral mucosa: a cautionary note. J. Oral Pathol. Med.,25: 60–64,1996.43. el-Deiry, W. S., Tokino, T., Waldman, T., Oliner, J. D., Velculescu,V. E., Burrell, M, Hill, D. E., Healy, E., Rees, J. L., Hamilton, S. R.,Kinzler, K. W., and Vogelstein, B. Topological control of p21WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res.,55:2910–2919, 1995.44. Nadal, A., Jares, P., Cazorla, M., Fernandez, P. L., SanJuan, X.,Hernandez, L., Pinyol, M., Aldea, M., Mallofre, C., Muntane, J.,

Traserra, J., Campo, E., and Cardesa, A. p21WAF1/Cip1 expressionis associated with cell differentiation but not with p53 mutations insquamous cell carcinomas of the larynx. J. Pathol.,183: 156 –163,1997.

45. Jung, J. M., Bruner, J. M., Ruan, S., Langford, L. A., Kyritsis, A. P.,Kobayashi, T., Levin, V. A., and Zhang, W. Increased levels ofp21WAF1/Cip1 in human brain tumors. Oncogene,11: 2021–2028,1995.

46. Barboule, N., Mazars, P., Baldin, V., Vidal, S., Jozan, S., Martel, P.,and Valette, A. Expression of p21WAF1/CIP1 is heterogeneous andunrelated to proliferation index in human ovarian carcinoma. Int. J.Cancer,63: 611–615, 1995.

47. DiGiuseppe, J. A., Redston, M. S., Yeo, C. J., Kern, S. E., andHruban, R. H. p53-independent expression of the cyclin-dependentkinase inhibitor p21 in pancreatic carcinoma. Am. J. Pathol,147:884–888, 1995.

48. el-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons,R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein,B. WAF1, a potential mediator of p53 tumor suppression. Cell,75:817–825, 1993.

49. Erber, R., Klein, W., Andl, T., Enders, C., Born, A. I., Conradt,C., Bartek, J., and Bosch, F. X. Aberrant p21(CIP1/WAF1) proteinaccumulation in head-and-neck cancer. Int. J. Cancer,74: 383–389,1997.

50. Awwad, S., Jaros, E., Somes, J., and Lunec, J. p53 overexpressionin head and neck carcinoma and radiotherapy results. Int. J. Radiat.Oncol. Biol. Phys.,34: 323–332, 1996.




1999;5:2455-2463. Clin Cancer Res M. Elizabeth Saunders, Robert MacKenzie, Rob Shipman, et al. RadiotherapyFull-Length Gene Sequencing and Results of Primary

Gene Mutations in Head and Neck Cancer:p53Patterns of

Updated version

http://clincancerres.aacrjournals.org/content/5/9/2455

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/5/9/2455.full#ref-list-1

This article cites 47 articles, 11 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/5/9/2455.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/5/9/2455To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/5/9/2455.full#ref-list-1

http://clincancerres.aacrjournals.org/content/5/9/2455.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



patterns of p53 gene mutations in head and neck cancer: full-the levels of p53 protein using ihc, on...

Documents