Clinical Cancer Research Meeting Calendar AACR Membership
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Clinical Cancer Research 13, 6099, October 15, 2007. doi: 10.1158/1078-0432.CCR-07-1369
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Huang, X.
Right arrow Articles by Partridge, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Huang, X.
Right arrow Articles by Partridge, M.

Imaging, Diagnosis, Prognosis

p53 Mutations in Deep Tissues Are More Strongly Associated with Recurrence than Mutation-Positive Mucosal Margins

Xiaohong Huang1, Stelios Pateromichelakis1, Allison Hills1, Martyn Sherriff2, Andrew Lyons1, John Langdon1, Edward Odell3, Peter Morgan3, John Harrison3 and Max Partridge1

Authors' Affiliations: 1 Maxillofacial Surgery, King's College Hospital, 2 Dental Biomaterial Science and Biomimetics, Guy's Hospital, and 3 Oral Pathology, Dental Institute, Guy's Hospital, King's College London, London, United Kingdom

Requests for reprints: Max Partridge, Maxillofacial Surgery, King's College Hospital, King's College London, Denmark Hill, London SE5 8RX, United Kingdom. Phone: 44-207-346-3474; Fax: 44-207-346-3754; E-mail: oralsurgery{at}partridgekcl.co.uk.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: Application of ultrasensitive diagnostics has shown that small numbers of p53 mutation-positive cells may signify the presence of residual tumor in histologically normal tissues after resection of squamous cell carcinomas arising in the head and neck area. To date, most studies in this area have focused on analysis of tissues at the mucosal aspect of the resection and highlighted the importance of molecular changes in the field with respect to the risk of recurrence.

Experimental Design: In the present investigation, we analyzed normal tissues from mucosal and deep surgical margins, referred to as "molecular margins," for the presence of the signature p53 mutation identified for each tumor.

Results: The p53 mutation status of these carcinomas did not correlate with clinical or histopathologic variables, but these mutations provided an excellent target for ultrasensitive analysis of margin status. We found that 11 of 16 (68%) of cases with histologically tumor-free (including 9 without dysplasia), but with p53 mutation-positive molecular margins, developed recurrence. The probability of developing local recurrence was significantly higher for the group with p53 mutation-positive margins when compared with the group with clear margins (P = 0.048) and more strongly associated with p53 mutation-positive deep molecular margins than mutation-positive mucosal molecular margins or positivity at both sites (P = 0.009).

Conclusions: This shows that although persistent mucosal fields may contribute to recurrence, clonal p53 mutations in deep tissues are an important cause of treatment failure, and molecular margins from both sites should be analyzed in future prospective series.

There has been only a modest improvement in the survival of cases treated for head and neck carcinoma over the last 20 years, in part because obtaining clear resection margins is difficult. Local or locoregional recurrence remains the most important cause of treatment failure, and there is a wealth of evidence showing that the risk of recurrence is increased when the resection margins are close as judged by light microscopy (1, 2). Based on this knowledge, the surgeon aims to resect a carcinoma with a 1-cm margin, but as margin retraction may be as high as 50% (3), the clearance as reported by the pathologist is often ≤5 mm.

Development of standardized criteria for reporting details of the pathologic assessment for head and neck carcinomas means that information about the closest mucosal and deep margins, for the sections examined by the pathologist, is now available. Having clear surgical margins is considered to be a good indicator of a favorable outcome. The United Kingdom guidelines designate a margin of >5 mm as being clear, 1 to 5 mm as close, and <1 mm as signifying likely incomplete excision (4). However, it is well recognized that margins of ≥5 mm may not be clear when the pattern of invasion is unfavorable (5). Anatomic constraints, together with the inherent difficulty in defining where the tumor margin lies by palpation, mean that many cases are resected with margins that are <5 mm (6). The presence of dysplasia at the resection margins is also included in the pathologist's report and provides additional prognostic information about the risk of recurrence (7) or development of a second primary tumor (8). The introduction of molecular analysis of surgical margins (9–14) has shown that morphologic information frequently underestimates the extent of spread of carcinoma or precancerous field effects. Follow-up of these cases has shown that "margins" harboring p53 mutations may be responsible for local or regional recurrence when they are pronounced clear after light microscopy (9–14), indicating that molecular assessment can augment the information provided by light microscopy.

At present, most molecular analysis is based on finding the same signature p53 gene mutation in a carcinoma and the normal tissues that remain at the edges of the defect after surgery. The strength of this approach is that p53 gene mutations are the most frequently detected aberration associated with head and neck carcinomas (15). However, although the presence of p53 mutations in the deep margins must signify carcinoma, if contamination has been avoided, the presence of the these mutations in mucosal margins may signify tumor, but these aberrations are also present in dysplasia and may be detected in oral mucosa that looks normal morphologically (10, 12, 16).

The present study was designed to extend this method of assessment and investigate whether the reliability of this assay for identifying cases likely to develop recurrence can be improved by analyzing more deep margins in addition to samples taken from the mucosal edge of the defect. In this report, we refer to the tissues that are forwarded for molecular analysis as "molecular margins." This is important as the term "surgical margin" frequently causes confusion as it may be wrongly assumed that the tissues examined are harvested from the edge of the resected carcinoma rather than the normal tissues that remain in the body after surgery. The picture has also become confused as margins taken from the excised surgical specimen have been included in some reports where molecular diagnostics have been applied (14).

In recent years, most effort has focused on determining the contribution of cells harboring mutant p53 in mucosal margins. The present report reveals that it is residual cancer in deep tissues that has the most adverse effect on outcome and shows that relying on identifying p53 mutation-positive mucosal fields as an additional diagnostic tool underestimates the extent of the problem. Based on this knowledge, we recommend that deep molecular margins as well as tissue from the mucosal edges of the defect are analyzed in future prospective studies.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
This study was approved by the Research Ethics Committees at Guy's and King's College Hospitals. Informed written consent was obtained from all participants. Immediately after tumor resection, the surgeon changed gloves and excised both mucosal and deep molecular margins from the tissues remaining at the edge of the defect using a new set of instruments. When the base of the resection was in continuity with the neck dissection and there were no true deep molecular margins, the tissues were excised from the lateral margins of the defect. On some occasions, T1 lesions were deemed to be too small to support removal of part of the tumor for molecular analysis, as there was a risk that there might be no tumor in the tissue processed for light microscopy. In addition, mucosal margins could not be taken when excising tumors that extended onto or arose on the vermillion border of the lip, as it was felt that removal of additional tissue might compromise the surgical repair.

We aimed to collect four to six molecular margins for each squamous carcinoma with an equal number of mucosal and deep samples. Each margin was divided into two halves. One part was snap frozen in isopentane for molecular analysis and the other was fixed in formalin and processed into paraffin for light microscopy. The pathologist removed a portion of each tumor from the resection specimen that was snap frozen and stored at –80°C until nucleic acids were extracted. Fifty-eight cases, with molecular margins assessed as being tumor-free by the pathologist after examination of the part of each sample sent for light microscopy, were forwarded for molecular analysis. These cases were followed for a minimum of 24 months to assess the effect of molecular detection of residual tumor on local recurrence (at the primary site) or locoregional recurrence (at the primary site and in the neck or in the neck only).

Molecular analysis. DNA was prepared from the frozen portion of the tumor and the molecular margins using DNA STAT-60 (Biogenesis) and RNA was back extracted according to the manufacturer's instructions. p53 cDNA was then synthesized from RNA and analyzed for the presence of a mutation in exons 4 to 10 (10). The primers used for PCR were 4F-5'-TGGAAACTACTTCCTGAAAAC-3', 6R-5'-ACACATGTAGTTGTAGTGGAT-3', 7F-5'-GACAGAAACACTTTTCGACAT-3', and 10R-5'-CAGCCAAAGAAGAAACCACTG-3'.

Subsequently, exons harboring p53 gene mutations were amplified from genomic DNA for each molecular margin analyzed by PCR; products were cloned into a bacteriophage vector and amplified in Escherichia coli. Each sample was plated in duplicate or triplicate for filter hybridization with oligonucleotide probes specific for the mutation identified for each tumor and the matched wild-type p53 sequence. For each margin analyzed, between 2,000 and 5,000 clones (of which ~25% contained a p53 DNA insert) were transferred to filters and each plaque assay included appropriate positive and negative controls. Hybridizations were carried out in Denhardt's buffer (5x SSC, 5x Denhardt's solution, 10x SDS) beginning at 11°C below the melting temperature of each oligonucleotide. Washing steps were 5x SSC at the hybridization temperature for 5 min followed by sequential washes, raising the temperature by 2°C in a stepwise manner until the negative control was clear as determined by phosphoimaging. The wash conditions were made more stringent by reducing the concentration of the SSC as appropriate. When examination of the filters indicated significant cross-hybridization between oligonucleotide probes recognizing mutant and wild-type p53 sequences, we incorporated mismatches into these probes to increase the specificity of the hybridization. The percentage of clonal (mutated) tumor cells in each sample was estimated by assuming that each cancer cell contained two copies of the mutant p53 allele and counting the number of plaques labeled with the mutant probe and dividing this by total number that hybridize with the wild-type (10).

All p53 mutation-positive tests were validated by reamplifying the positive plaques and the test was repeated for cases scored as having <0.1% mutant p53 clones in a molecular margin with a new aliquot of gDNA and cDNA (prepared from the RNA that was back extracted during preparation of DNA). This step is considered to ensure that a tumor-positive result is not a false-positive test due to amplifying DNA from dead tumor cells. These stringent validation steps were also done for those cases, scored as having p53 mutations in molecular margins that did not develop locoregional recurrence during the study period.

Statistical analysis. Data were analyzed using either Stata Release 9.2 or StatXact 3. Significance was predetermined at {alpha} = 0.05. Exact nonparametric inference was used to allow for the small sample size. Single ordered data were analyzed using the Kruskal-Wallis test and double ordered using the Jonckheere-Terpstra test. The Kaplan-Meier test was used for survival analysis, and Sidak's test was used to evaluate multiple comparisons. Logistic regression, in conjunction with Bayesian information criterion, was used to select predictors for defined outcomes (17, 18).


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Relationship between p53 mutation status and clinical and pathologic features. The clinical and pathologic features of 58 consecutive cases with tumor-free resection margins by light microscopy forwarded for analysis are shown in Table 1 . Thirty-one cases received postoperative radiotherapy, and the follow-up time for the study ranged from 5 to 62 months with the median being 26 months. All surviving cases were followed for at least 24 months. At the time of assessment, 26 cases had developed local or locoregional recurrence, and there were 2 distant recurrence and 2 second tumors (Table 1). We searched for relationships between p53 status and clinicopathologic features, including the number of tumor-positive nodes, T or N stage, the presence of extracapsular spread, perineural or vascular invasion, pushing or invasive margins, tumor grade, the presence of dysplasia at the resection margins, and the clearance at the resection margins, but none was found. The only significant relationship was between morphologic variables in that infiltrative tumor margins were associated with neural invasion (P = 0.008).


View this table:
[in this window]
[in a new window]

  		Table 1. Clinicopathologic details for 58 cases analyzed with respect to tumor p53 gene status

 
Analysis of mucosal and deep molecular margins. Twenty-five p53 mutations could be used to develop plaque assays. The remaining five cases could not be interrogated using this method as it was not possible to design oligonucleotides that could reliably distinguish between the mutant and wild-type sequences. These mutations included two single base pair deletions (codons 326.2 and 376.2) and three point mutations (176.2 G-T, 240.2 G-A, and 175.1 G-A).

Forty mucosal and 46 deep molecular margins were analyzed for 25 cases and 16 (64%) showed p53 mutations in one or more margins examined after all validation steps were completed. Overall, we found that an assay yielding more than 2 of 1,000 positive plaques was always confirmed when the test was repeated with cDNA. However, there is no lower number of positive plaques that can be assessed as likely representing a false-positive result, as a single tumor-positive colony among 4,000 plaques (0.01% positive clones) was verified by stabbing in an earlier study (12).

We did find some discordant results when both gDNA and cDNA were analyzed in that, of three margins that were positive with gDNA for case 1, only two were positive with cDNA and four of six were positive with both nucleic acids for case 7. However, all cases scored as having p53 mutation-positive mucosal or deep molecular margins with gDNA were confirmed as such after analysis of cDNA.

Five of the 16 cases had mucosal molecular margins that were scored as p53 mutation-positive, 7 had tumor-positive deep molecular margins, and 4 cases were positive at both sites. The percentage of p53 mutant clones ranged from 0.1 to 17 for the mucosal margins and from 0.1 to 6.75 for the deep. Six cases had a single p53 mutation-positive margin, and 10 had multiple positive margins, with 6 of this group being positive at all sites examined. Seven of these cases had morphologic evidence of dysplasia at the resection margins. However, only one case had dysplasia in the half of the molecular mucosal margin examined by the pathologist and no malignant cells were identified, highlighting that p53 mutations are often found in tumor adjacent tissue in the absence of morphologic changes. The key clinicopathologic characteristics of the cases examined with the molecular diagnostic and the details about margin status are summarized in Table 2 .


View this table:
[in this window]
[in a new window]

  		Table 2. Clinical details, including information about the presence of dysplasia at the resection and molecular margins, the closest surgical margins, and outcome with the results following application of the p53 phage plaque assay for mucosal and deep molecular margins

 
Relationship between clinicopathologic features and p53 mutation-positive molecular margins. We looked for evidence of a relationship between clinical (age, sex, tumor site and stage, and T or N classification) or histologic features (including tumor grade, type of margin, and nerve or vascular invasion) and p53 mutation-positive molecular margins but none was found. In addition, we considered the sites from which the molecular margins were excised, as previously we detected more residual malignant cells in molecular margins harvested from the tissues remaining at the posterior aspect of the surgical defect (10), but recurrences occurred at all sites in the present series.

Relationships between molecular analysis of margins and local recurrence. We found that a very high number of cases with p53 mutation-positive molecular margins developed local (11 of 16, 69%) or locoregional recurrence (12 of 16, 75%). During the follow-up period, an adenocarcinoma was discovered in the femur for one case, but there were no second primary carcinomas in the head and neck, likely reflecting the application of stringent criteria for diagnosing second tumors (19).

The effect of p53 mutation-positive molecular margins on local and locoregional recurrence was assessed by Kaplan-Meier estimates and log-rank tests. This analysis showed that there was an increased probability of developing local or locoregional recurrence for the group with mutation-positive molecular margins (P = 0.048 and 0.06, respectively). Overall, there was a 37% increase in local recurrence when the molecular margins were scored as p53 mutation positive (Fig. 1A ) and a significant difference between the local recurrence rates when cases with molecular margins scored as clear or p53 mutation-positive at the mucosal or deep aspect, or at both sites, were compared (P = 0.009; Fig. 1B). The difference was most significant when clear and tumor-positive deep margins were compared. A similar effect was found for p53 mutation-positive deep molecular margins and development of locoregional recurrence (P = 0.027; data not shown).


Figure 1
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. A, Kaplan-Meier estimates for no local recurrence with respect to the presence or absence of p53 mutations in the molecular margins examined. B, Kaplan-Meier estimates for no local recurrence with respect to clear molecular margins, p53 mutation-positive mucosal molecular margins, p53 mutation-positive deep molecular margins, or positivity at both sites (B).

 
Examination of the present cohort did not find any relationship between a p53 mutation-positive molecular margin and close (<5 or <3 mm) or infiltrative margins as measured by light microscopy. However, our data highlight the extent of tumor spreading at the molecular level, as 7 of 16 cases with p53 mutations in the molecular margins had resection margins that were clear at ≥5 mm after morphologic examination. p53 mutation-positive mucosal molecular margins were also found when the clearance reported by the pathologist was 9 mm for the mucosal aspect and 11 mm for the deep.

Concordance between p53 mutation-positive molecular margins, development of recurrence, and the effect of radiotherapy. The finding of p53 mutation-positive molecular margins correctly identified 12 cases developing local or locoregional recurrence and 6 scored as tumor-negative that did not recur. However, 3 cases with p53 mutation-negative molecular margins developed recurrence and 5 p53 mutation-positive cases did not recur. The sensitivity, specificity, positive predictive value (the proportion of patients with a positive test result who develop recurrence), and negative predictive value (the proportion of patients with a positive test result who develop recurrence) are summarized in Table 3 . These data show that the current tissue sampling technique fails to detect all cases at risk of recurrence, although it is to be anticipated that the finding of p53 mutations will not necessarily translate into risk of recurrence, as some cells with mutant DNA will be eliminated by radiotherapy. Eleven of 16 cases with p53 mutation-positive molecular margins received postoperative radiotherapy but 8 developed locoregional recurrence, highlighting the fact that this treatment modality cannot be relied on to destroy any remaining malignant cells.


View this table:
[in this window]
[in a new window]

  		Table 3. The sensitivity, specificity, and positive and negative predictive values for the p53 phage plaque assays for local and locoregional recurrence

 
Relationship between clinicopathologic features and development of recurrence. A combination of factors is associated with adverse outcome, and we also looked for evidence of a relationship between clinicopathologic factors of the cases and development of local or locoregional recurrence. Nodal status (P = 0.002) and the presence of dysplasia at the resection margins (for 20 of 58 cases, P = 0.023) were significantly associated with development of local or locoregional recurrence. However, application of the molecular diagnostic identified cases without dysplasia at the resection or molecular margins (16 and 22 cases, respectively; Table 4 ) but with p53 mutation-positive molecular margins (9 and 14 cases) developing local recurrence, highlighting the additional prognostic information provided by the molecular test in the absence of morphologic changes at the margins examined.


View this table:
[in this window]
[in a new window]

  		Table 4. Development of local recurrence for cases with and without evidence of dysplasia at the resection or molecular margins and p53 mutation-positive molecular margins highlighting the additional information provided by the molecular test when there is no evidence of dysplasia

 
Combining molecular and morphology information to improve prediction of recurrence. We identified cases with tumor-negative margins that developed recurrence and looked for ways of combining molecular and morphologic information to improve the reliability of the p53-based molecular diagnostic to identify good prognosis patients. We found that when we considered information about tumor-free molecular margin status together with the clearance at the margins as measured by light microscopy, cases with clear molecular margins that were also tumor-free at ≥5 mm by light microscopy did not recur. This suggests that combining molecular and morphologic information may be useful when searching for ways of identifying the cohort of cases who may be cured by surgery and postoperative radiotherapy more precisely.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
It is now generally accepted that the tumor-node-metastasis system together with the pathologist's report outlining the pattern of tumor growth provides the most reliable information about prognosis that can be obtained by light microscopy. However, the challenge is to develop molecular markers that can provide more accurate predictions about the risk of recurrence and establish the most appropriate samples that should be tested in a routine diagnostic setting. In the present study, we found a high percentage of cases with histologically tumor-free resection margins, but p53 mutation-positive molecular margins developed local or locoregional recurrence and that analyzing deep margins was most useful in terms of predicting risk of recurrence.

Overall, we found that identical p53 mutations were present in the tumor and molecular margins for 16 of 25 cases adding to the literature showing that spread of carcinoma may be much greater than previously realized (9–13). A high number of cases with p53 mutation-positive molecular margins developed local (68%) or locoregional (75%) recurrence. The percentage of cases with p53 mutation-positive molecular margins developing recurrence is higher than previous reports, which indicated that local or locoregional recurrence developed in 18% to 50% of cases (9–13). The high recurrence rate observed in the present study is not due to inclusion of more advanced cases, as the percentage of early (T1 and T2, 26%) and advanced (T3 and T4, 74%) tumors forwarded for molecular analysis is broadly similar to the stage distribution of unselected cases treated by the same authors (20).

Assays for detection of residual cancer aim to analyze additional tissues from the mucosal and deep aspects of the defect after surgery. Previous reports have generally included many more mucosal margins as it is easier to remove additional tissues from the mucosal edges and there may be no inferior margins when the tumor and neck dissection are removed in continuity. However, tissues from the lateral aspects of the defect can readily be sampled. Although the finding of p53 mutations in a mucosal margin may not necessarily signify tumor, the presence of mutant p53 gDNA or cDNA in deep tissues must represent spread of carcinoma, providing that all appropriate steps have been taken to prevent transfer of malignant cells from the tumor to normal tissues (10, 12, 13). We were able to analyze the effect of p53 mutation-positive mucosal and deep molecular margins on development of recurrence separately, as the margins sampled included equal numbers of mucosal and deep tissues. When the effect of p53 mutation-positive mucosal or deep molecular margins on development of local of locoregional recurrence was analyzed separately, we found that p53 mutation-positive deep molecular margins have the most adverse effect on outcome.

There has been considerable discussion in the literature about the concept of mucosal fields based on the accumulating evidence that the extent spread of altered cells is much greater than previously realized (21–23). These studies have established conclusively that large areas of the oral mucosa may harbor the genetic mutations associated with tumors and the concept of a "field" aptly describes the location of these aberrations throughout the superficial tissues. However, the rate of local recurrence associated with p53 mutations in mucosal margins is low (10, 12, 13). Firm evidence that the epithelial cells with mutated p53 that are detected with the phage plaque assay give rise to second tumors is also lacking, although microsatellite analysis has shown that tumors and the surrounding tissues may harbor the same aberrations (24).

We were surprised to find that only four of nine cases with the same p53 mutation in the tumor and the mucosal molecular margins developed recurrence. Several factors may contribute to this discrepancy. The presence of a p53 mutation in a mucosal margin may indicate that malignant cells remain in the field, but these aberrations are also associated with dysplasia and may be present in histologically normal, inflamed mucosa or epithelia showing other morphologic changes (10, 25, 26), reducing the usefulness of these mutations for identifying rare malignant cells against a background of normal. It is to be anticipated that some cells harboring p53 mutations may be eliminated by radiotherapy. In addition, as the oral mucosa is frequently exposed to tobacco-derived carcinogens, it is likely that many p53 mutations will be in transient amplifying cells that will be lost on differentiation (27). However, some p53 mutation-positive mucosal patches may contain slowly cycling stem cells that harbor these mutations such that their progeny may be responsible for the persistence of patches of keratinocytes harboring these mutations. Over time, these abnormal clones may acquire additional aberrations, leading to development of a focus of recurrent tumor that will by definition harbor the same p53 mutation.

The possibility that discordant results arise due to contamination of margins with DNA from dead tumor cells has also been suggested based on the finding that not all tumor-positive mucosal margins were confirmed as such when the test was repeated with cDNA (12). In the present series, all test results yielding <0.1% positive clones were based on analysis of gDNA and cDNA, and we are confident that the low incidence of recurrence associated with a positive mucosal margin in the present series is not due to technical issues. The finding that all cases with p53 mutation-positive deep molecular margins developed local or locoregional recurrence reinforces this statement. However, the effect of radiotherapy and the likely existence of p53 gene mutations in epithelial cells with different proliferative potential means that establishing the sensitivity and specificity of any molecular test for detection of residual carcinoma provides only a guide to the utility of this approach. Nevertheless, a sensitivity of 70% for the p53 phage plaque assay provides a measure that can be considered as a "gold standard" against which the alternative methods that are in development (28, 29) can be assessed.

We found that the molecular analysis identified p53 mutation-positive molecular margins for cases with clear or close margins as defined by light microscopy. The current United Kingdom standard for establishing clear resection margins as part of the data set for histopathology reports is >5 mm (4). In the present series, 7 of the 16 cases with p53 mutations in molecular margins, but resection margins that were tumor-free at ≥5 mm after morphologic examination, developed locoregional recurrence. These mutations were also found when the mucosal and deep margins were tumor-free at 9 and 11 mm, respectively.

It is well recognized that examination of resection margins by light microscopy provides only a guide to margin status, and our data support the view that even 5 mm clearance may not be sufficient if there is evidence of an infiltrative pattern of spread (5), manifested in the present study by molecular detection of p53 mutations in the molecular margins. The finding of p53 mutations in the absence of dysplasia also reinforces this view. Thus, although the accuracy of morphologic examination can be improved by examining deeper levels and additional sections when the resection margins are found to be close, molecular approaches offer an attractive method for screening for evidence of p53 gene mutations in tissues beyond the resection margins. In the present study, we did not find a significant relationship between an infiltrative pattern of growth and neural or vascular invasion and p53 mutation-positive molecular margins, but these characteristics were frequently associated with mutation-positive molecular margins.

The concept of microtumor deposits, as detected by light microscopy, is generally accepted and several reports have provided molecular evidence of extensive tumor spreading of head and neck carcinomas (10–14, 19, 21, 22). This is the first study to show that clonal p53 mutations in deep molecular margins have a very adverse effect on outcome. Thus, it seems timely to suggest that more emphasis is given to the concept of DNA-based detection of carcinoma in deep tissues. Previous reports based on light microscopy have shown that the deep margin is involved more than the mucosal (30) and have highlighted the importance of close deep resection margins. At present, it is not known whether malignant cells spread as single entities or as clusters or whether the rare p53 mutations detected in deep tissues are in carcinoma cells or those that have undergone epithelial-mesenchymal transition. Precancerous cells harboring p53 mutations may also migrate through the epithelia or be spread via the vasculature or lymphatics (19, 31, 32).

Our findings suggest that it may also be advantageous to combine molecular and conventional morphologic information to identify cases with a good prognosis. We found that, when there was no molecular evidence of p53 mutations in mucosal or deep molecular margins, and when the resection margins were tumor-free at 5 mm by light microscopy, no case recurred and the utility of combining conventional and emerging molecular approaches can be tested in the next phase of the clinical study.

The significance of these results for patient management must also be considered. In the present series, most cases with p53 mutation-positive deep molecular margins treated by surgery and postoperative radiotherapy developed recurrence, suggesting that these tumors contain populations that are radioresistant. A larger study will be required to conclusively establish the effect of postoperative radiotherapy, but these data suggest that it may be better to wait for the result of molecular analysis before repairing a defect with a microvascular flap to avoid burying residual carcinoma. Alternative treatment regimens incorporating tumor-targeted agents, such as cetuximab, Iressa, and gefitinib, or antiangiogenic agents (33) are showing benefit when given with radiotherapy or as part of combination treatment regimens, and it is to be hoped that these advances will also provide a step forward in the management of head and neck cancer.


   Footnotes
 
Grant support: British Association of Oral and Maxillofacial Surgeons and the Special Trustees of Guy's and St. Thomas' Hospitals.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 6/ 1/07; revised 8/ 3/07; accepted 8/29/07.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Loree TR, Strong EW. Significance of positive margins in oral cavity squamous cell carcinoma. Am J Surg 1990;60:410–4.
  2. Spiro RH, Guillamondegui O, Jr., Paulino AF, Huvos AG. Pattern of invasion and margin assessment in patients with oral tongue cancer. Head Neck 1999;21:408–13.[CrossRef][Medline]
  3. Johnson RE, Sigman JD, Funk GF. Quantification of surgical margin shrinkage in the oral cavity. Head Neck 1997;19:281–6.[CrossRef][Medline]
  4. Standards and datasets for reporting cancers. Datasets for histopathology reports on head and neck carcinomas and salivary neoplasms. 2nd ed. London: The Royal College of Pathologists; 2005.
  5. Woolgar JA. Histopathological prognosticators in oral and oropharyngeal squamous cell carcinoma. Oral Oncol 2006;42:229–39.[CrossRef][Medline]
  6. Woolgar JA, Rogers S, West CR, Errington RD, Brown JS, Vaughan ED. Survival and patterns of recurrence in 200 oral cancer patients treated by radical surgery and neck dissection. Oral Oncol 1999;35:257–65.[CrossRef][Medline]
  7. Weijers M, Snow GB, Bezemer PD, van der Wal JE, van der Waal I. The clinical relevance of epithelial dysplasia in the surgical margins of tongue and floor of the mouth squamous cell carcinoma: an analysis of 37 patients. J Oral Pathol Med 2002;31:11–5.[CrossRef][Medline]
  8. Slootweg PJ, Hordijk GJ, Schade Y, Van Es RJJ, Koole R. Treatment failure and margin status in head and neck cancer. A critical view on the potential value of molecular pathology. Oral Oncol 2002;38:500–3.[CrossRef][Medline]
  9. Brennan JA, Mao L, Hruban RH, et al. Molecular assessment of histopathological staging in squamous-cell carcinoma of the head and neck. N Engl J Med 1995;332:429–35.[Abstract/Free Full Text]
  10. Partridge M, Li S-R, Pateromichelakis S, et al. Detection of minimal residual cancer to investigate why oral tumors recur despite seemingly adequate treatment. Clin Cancer Res 2000;60:2718–25.
  11. Sardi I, Franchi A, Ferriero G, et al. Prediction of recurrence by microsatellite analysis in head and neck cancer. Genes Chromosomes Cancer 2000;29:201–6.[CrossRef][Medline]
  12. van Houten VM, van den Brekel MW, Kummer JA, et al. Mutated p53 as a molecular marker for the diagnosis of head and neck cancer. J Pathol 2002;198:476–86.[CrossRef][Medline]
  13. van Hourten VMM, Leemans CR, Kummer JA, et al. Molecular diagnosis of surgical margins and local recurrence in head and neck cancer patients: a prospective study. Clin Cancer Res 2004;10:3614–20.[Abstract/Free Full Text]
  14. Temam S, Casiraghi O, Lahaye JB, et al. Tetranucleotide microsatellite instability in surgical margins for prediction of local recurrence of head and neck squamous cell carcinoma. Clin Cancer Res 2004;10:4022–8.[Abstract/Free Full Text]
  15. Partridge M, Gaballah K, Huang X. Molecular markers for diagnosis and prognosis. Cancer Metastasis Rev 2005;24:71–85.[CrossRef][Medline]
  16. Waridel F, Estreicher A, Bron L, et al. Field cancerisation and polyclonal p53 mutation in the upper aerodigestive tract. Oncogene 1997;14:163–9.[CrossRef][Medline]
  17. Raftery A. Bayesian model selection in social research. Sociol Methodol 1995;25:111–63.[Medline]
  18. Schwarz G. Estimating the dimension of a model. Ann Stat 1978;6:461–4.
  19. Pateromichelakis S, Farahini M, Phillips E, Partridge M. Molecular analysis of paired tumours: time to start treating the field. Oral Oncol 2005;41:916–26.[CrossRef][Medline]
  20. Evans SJW, Langdon JD, Rapidis, Johnson NW. Prognostic significance of STNMP and velocity of tumour growth in oral cancer. Cancer 1982;49:773–6.[CrossRef][Medline]
  21. Partridge M, Pateromichelakis S, Phillips E, Emilion GG, A'Hern RP, Langdon JD. A case-control study confirms that microsatellite assay can identify patients at risk of developing oral SCC within a field of cancerization. Cancer Res 2000;60:3893–8.[Abstract/Free Full Text]
  22. Braakhuis BJM, Tabor MP, Leemans CR, van der Waal I, Snow GB, Brakenhoff RH. Second primary tumors and field cancerization in oral and oropharyngeal cancer: molecular techniques provide new insights and definitions. Head Neck 2002;24:198–206.[CrossRef][Medline]
  23. Bedi GC, Westra WH, Gabrielson E, Koch W, Sidransky D. Multiple head and neck tumors: evidence for a common clonal origin. Clin Cancer Res 1996;56:2484–7.
  24. Tabor MP, Brakenhoff RH, Ruijter-Schippers HJ, et al. Multiple head and neck tumors frequently originate from single preneoplastic lesions. Am J Pathol 2002;161:1051–60.[Abstract/Free Full Text]
  25. Hussain SP, Amstad P, Raja K, et al. Increased p53 mutation load in noncancerous colon tissue from ulcerative colitis: a cancer-prone chronic inflammatory disease. Cancer Res 2000;60:3333–7.[Abstract/Free Full Text]
  26. Kashiwazaki H, Tonoki H, Tada M, et al. High frequency of p53 mutations in human oral epithelial dysplasia and primary squamous cell carcinoma detected by yeast functional assay. Oncogene 1997;15:2667–74.[CrossRef][Medline]
  27. Barrandon Y, Green H. Three clonal types of keratinocyte with different capacities for multiplication. Proc Natl Acad Sci U S A 1987;84:2302–6.[Abstract/Free Full Text]
  28. Harden SV, Thomas DC, Benoit N, et al. Real-time Gap ligase chain reaction: a rapid semiquantitative assay for detecting p53 mutation at low levels in surgical margins and lymph nodes from resected lung and head and neck tumors. Clin Cancer Res 2004;10:2379–85.[Abstract/Free Full Text]
  29. Shi C, Eshelman SH, Jones D, et al. LigAmp for sensitive detection of single-nucleotide differences. Nat Methods 2004;2:1–7.
  30. Woolgar JA, Triantafyllou A. A histopathological appraisal of surgical margins in oral and oropharyngeal cancer resection specimens. Oral Oncol 2005;41:1034–43.[CrossRef][Medline]
  31. Braakhuis BJM, Leemans CR, Brakenhoff RH. Expanding fields of genetically altered cells in head and neck squamous carcinogenesis. Semin Cancer Biol 2005;15:113–20.[CrossRef][Medline]
  32. Partridge M, Brakenhoff RH, Phillips E, et al. Detection of rare disseminated tumor cells identifies head and neck cancer patients at risk of treatment failure. Clin Cancer Res 2003;9:5287–94.[Abstract/Free Full Text]
  33. Giaccone G. Clinical impact of novel treatment strategies. Oncogene 2002;21:6970–81.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Huang, X.
Right arrow Articles by Partridge, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Huang, X.
Right arrow Articles by Partridge, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online