|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology p53 Gene Status as a Predictor of Tumor Response to Induction Chemotherapy of Patients With Locoregionally Advanced Squamous Cell Carcinomas of the Head and NeckFrom the Service d’Oto-Rhino-Laryngologie et de Chirurgie de la Face et du Cou, Antenne de Biostatistiques, L’Institut National de la Santé et de la Recherche Médicale (INSERM) U444, Laboratoires de Génétique Moléculaire, d’Histologie et de Biologie Tumorale, and Central d’Anatomie et de Cytologie Pathologiques, Hôpital Tenon, Paris, France. Address reprint requests to Pierre Fouret, MD, Service d’Anatomie Pathologique, Hôpital Tenon, 75970 Paris Cedex 20, France; email fouret{at}infobiogen.fr
PURPOSE: To determine whether p53 gene status predicts tumor responses to platinum- and fluorouracil-based induction chemotherapy in locoregionally advanced squamous cell carcinomas of the head and neck. PATIENTS AND METHODS: Tumor responses of 105 patients were measured at the primary tumor site. Objective response and major response were defined by a 50% and 80% reduction in tumor size, respectively. All coding parts of p53 gene were directly sequenced. p53 expression in tumor cells was determined by immunohistochemistry. Human papillomavirus infection was detected by polymerase chain reaction. Odd ratios were adjusted by stepwise logistic regression analysis. RESULTS: p53 mutations, p53 expression, and tumor stage were sufficient to explain the variation in tumor responses to chemotherapy in multivariate models. p53 mutation was the only variable to significantly predict objective response (odds ratio, 0.23; 95% confidence interval, 0.10 to 0.57; P = .002) and was the strongest predictor of major response (odds ratio, 0.29; 95% confidence interval, 0.11 to 0.74; P = .006). p53 expression (odds ratio, 0.39; 95% confidence interval, 0.16 to 0.98) and tumor stage (odds ratio, 0.31; 95% confidence interval, 0.10 to 0.96) also predicted major response. Specific mutations (contact mutations) accounted for much of the reduction in the risk of major response associated with overall mutations. In complementary analyses, p53 expression was weakly predictive of major response in the subgroup with wild-type p53, and p53 mutations also predicted histologic response. CONCLUSION: p53 gene mutations are strongly associated with a poor risk of both objective and major responses to chemotherapy. Contact mutations are associated with the lowest risk of major response to chemotherapy.
THE ROLE OF CHEMOTHERAPY in squamous cell carcinomas of the head and neck (SCCHNs) has evolved in the last decades from being a palliative treatment to an essential component of multimodality therapy of locoregionally advanced tumors.1 The association of cisplatin and fluorouracil produces high tumor response rates when administered before surgery or radiation therapy.2,3 A major regression of tumor after two cycles of induction chemotherapy tends to predict a complete response to subsequent radiation therapy and is therefore used to select patients for definitive radiation instead of aggressive surgery. The benefit of this alternative to total laryngectomy is organ preservation, as demonstrated in the trials conducted by the Department of Veterans Affairs for laryngeal tumors4 and by the European Organization for Research and Treatment of Cancer for hypopharyngeal tumors.5 Complete responders to induction chemotherapy enjoy a survival advantage.6 However, most randomized trials have failed to demonstrate that the chemotherapy yields significant benefit either in survival or locoregional control,7 although in one study induction chemotherapy has been shown to improve survival in patients with unresectable disease.8 It has been speculated that chemotherapy only benefits patients with a less aggressive disease.9 Alternatively, this suggests that chemotherapy may benefit responders but may be detrimental to others. There is the risk that tumors progress when the patient is undergoing chemotherapy. Nonresponders who refuse salvage surgery have a poor prognosis.10 Detection of persistent disease or locoregional recurrence after treatment by chemotherapy and radiation is difficult in extensive fibrous tissue.11 Salvage surgical treatment is a challenging procedure with high rates of major wound complications.12 In this context, the identification of predictors of tumor response would allow a more rational selection of patients for standard induction chemotherapy. Future trials of new, potentially toxic, combination therapy should also benefit from selection of patients. At present, only the clinical tumor staging parameters seem significantly linked to tumor responses,13 although clinical staging often fails to predict individual response to treatment. The histologic growth pattern also seems related to responses to chemotherapy,14 although this finding has not been widely explored. Recent data indicate that thymidylate synthetase expression15 or tumoral reduced folates16 may help to identify patients who are more likely to respond to fluorouracil. Research on drug resistance has discovered many possible mechanisms.17 A major progress has been the realization that oncogenes that induce tumor growth also make tumor cells more sensitive than their normal counterparts to genotoxic agents such as chemotherapy or radiotherapy.18 Reciprocally, inactivation of tumor suppressor genes, for which tumors select to avoid oncogene-induced apoptosis, may be responsible for the resistance to genotoxic DNA damage.19,20 p53, one of the most frequently mutated antioncogenes, is required for the apoptotic response of transformed fibroblasts to several DNA-damaging agents, including radiotherapy and doxorubicin.21 Introduction of the p53 gene in p53-deficient SCCHN lines can induce apoptosis22 and sensitize cells to genotoxic treatment.23 Recent trials also show that adenoviral vector delivery of the wild-type p53 gene is associated with objective antitumor activity.24 p53 mutations predict locoregional failure in patient with SCCHN treated with radiotherapy25 and have been linked to poor responses to preoperative chemoradiotherapy of patients with esophageal carcinomas.26 Although there is a large amount of data concerning p53 gene status in SCCHN, no study has specifically addressed the question whether p53 gene status predicts tumor responses to platinum- and fluorouracil-based induction chemotherapy. For this purpose, we studied p53 gene status by directly sequencing all coding parts of the p53 gene in a retrospective series of 107 cases of locoregionally advanced SCCHN that had been treated by induction chemotherapy with platinum and fluorouracil. We also studied p53 expression by immunohistochemistry because p53 expression is often considered as a surrogate marker for mutation. We also determined human papillomavirus (HPV) infection because viral gene products may inactivate the p53 protein.27
Patients From September 1992 to December 1995, 276 cases of SCCHN (excluding nasopharyngeal tumors) were newly diagnosed in the Service d’Oto-Rhino-Laryngologie et de Chirurgie de la Face et du Cou at our institution (Hôpital Tenon, Paris, France). Of these 276 cases, 130 were stage III or IV tumors. Twenty-three cases with a poor performance status were not treated by induction chemotherapy. The remaining 107 patients with stage III or IV SCCHN were treated by induction chemotherapy and were all included in the present study. The study was approved by the institutional ethical committee and complied with current French laws. Patient characteristics were retrieved from the records. Histologic growth patterns were reviewed by two experienced pathologists (P.C. and P.F.) following published criteria.14
Treatment Plan Chemotherapy was administered according to a modified Al Sarraf protocol, which is used at several institutions in France. All patients had three planned cycles of chemotherapy. Each cycle consisted of cisplatin (20 mg/m2/d) administered as a continuous intravenous 24-hour infusion, simultaneously with a continuous infusion of fluorouracil (1,000 mg/m2/d) for 4 days. Chemotherapy was withdrawn in cases of tumor progression or in patients who did not respond after two cycles. Patients who stopped treatment because of toxicity were included in the study. Twenty-two patients underwent two cycles of chemotherapy, and two patients underwent only one cycle. The final response was estimated from the last clinical and endoscopic evaluation, which was performed under general anesthesia 2 weeks after the last cycle. Tumor shrinkage was evaluated as the decrease in the sum of the product of the largest perpendicular diameters of the measurable lesions at the primary tumor site. An objective response was defined by tumor shrinkage of more than 50% of initial tumor size. A major response was defined by a tumor shrinkage of more than 80% of initial tumor size. Major responders with laryngeal or hypopharyngeal tumors were treated by exclusive radiotherapy. No randomized trial has concluded that patients with oropharyngeal tumors can be treated by chemotherapy followed by radiotherapy alone, although some data suggest that a conservative approach is feasible in these cases.2,28 For oropharyngeal tumors, major responders to chemotherapy were proposed limited surgery of tumor remnants followed by radiotherapy, which was the standard treatment in these cases. Radiotherapy was performed at the same institution. The dosage was 70 Gy for radiotherapy alone and 45 Gy for radiotherapy after surgery. The dosage was increased to 70 Gy in tumors with involved margins.
DNA Samples For sequencing analysis of the p53 gene, three 8-µm unstained sections were deparaffinized and dehydrated just before microdissection. Using a 3-µm stained section as reference, tumor tissue was microdissected with surgical blades under a binocular lens at a 50x magnification. Samples enriched in tumor cells were placed into microfuge tubes containing 100 µL digestion buffer containing 200 µg/mL proteinase K and incubated overnight at 37°C. The proteinase K was inactivated by heating at 95°C for 10 minutes. For detection of HPV, one unstained 4-µm section from which excess paraffin had been removed was directly placed into digestion buffer and processed as previously described. The processing of slides, the preparation and storage of DNA samples, and the polymerase chain reactions (PCRs) were performed in other rooms than those in which PCR products were manipulated. DNA samples from unrelated tissues were used as controls.
Sequencing of the p53 Gene • Exons 2 and 3, first round: 5'CCCCACTTTTCCTCTTGC3', 5'GGACAAGGGTTGGGCTGG3'; second round: 5'CAGCCAGACT- GCCTTCCG3', 5'GGACAAGGGTTGGGCTGG3' • Exon 4, first round: 5'CCTCTGACTGCTCTTTTCAC3', 5'CGTGTATTCCTTGGCTTT3'; second round: 5'CCTCTGACTGCTCT- TTTCAC3', 5'TCTCATGGAAGCCAGCCCCT3' • Exon 10, first round: 5'TGTATAGGTACTTGAAGTGC3', 5'ACGTGGAGGCAAGAATGTGG3'; second round: 5'GAACCAT-CTTTTAACTCAGG3', 5'TGAGAATGGAATCCTATGGC3' • Exon 11, first round: 5'CATTGGTCAGGGAAAAGGGG3', 5'GCCTGCACTGGTGTTTTGTT3'; second round: 5'TCATGTGATGTCATCTCTCC3', 5'GCCATTTTGGGTTTTGGGTC3' The PCR products purified using Qiagen columns (Diagen, Dusseldorf, Germany) according to manufacturer’s instructions were sequenced using fresh internal primers and fluorescent dye–labeled dideoxynucleotides (Dye-Terminator sequencing kit; Perkin Elmer, Norwalk, CT). Fluorograms collected with a DNA sequencer (model 310; Applied Biosystems, Foster City, CA) were analyzed using the Sequence Navigator software (Applied Biosystems) to identify mutations by comparison with fluorograms from wild-type samples. Mutations were identified and confirmed in both strands in two independent amplification and sequencing experiments. Neutral mutations and known polymorphisms were not counted as mutations.31
HPV Detection
Immunohistochemistry Specimens were blindly evaluated for p53 staining by two pathologists (P.F. and S.T.). Intensity of staining and percentage of positive cells were recorded. Staining in less than 5% of tumor cells was scored as negative. Previous experiments had demonstrated that a good agreement between different observers could be obtained using these criteria.34 All series included immunoreactive breast carcinoma specimens as p53-positive external controls. Negative controls included replacement of the monoclonal antibody with a mouse myeloma antibody of the same subclass.
Statistical Analysis Predictor variables with a P value less than .25 in the univariate analyses were submitted to multivariate analysis with response to chemotherapy as the dependent variable. BMDP software (SPSS, Chicago, IL) was used to develop the multivariate stepwise logistic-regression models with forward selection to obtain the smallest number of explanatory variables that provided a well-fitting model. The Hosmer-Lemeshow goodness-of-fit criterion was used to determine the adequacy of the model. The relative risk to respond to chemotherapy was estimated by adjusted odds ratios with 95% confidence intervals. All reported P values were two-sided. A value of .05 was chosen to indicate statistical significance.
Forty tumors (37%) harbored one mutation (Table 1). Thirty-three (82.5%) mutations were localized in exons 5 to 9. Six mutations (15%) were localized in exon 4. Nineteen mutations (47.5%) were missense mutations. Two mutations (5%) were inframe deletions. The remaining 19 mutations (47.5%) included one frameshift insertion, five frameshift deletions (12.5%), seven stop mutations (17.5%), and six splice junction mutations (15%). Point mutations were predominantly G/C to A/T and A/T to G/C transitions (11 cases) or G/C to T/A transversions (nine cases).
There was perfect concordance between the type of mutation and p53 expression as detected by immunohistochemistry. Every tumor that contained a missense mutation or an inframe deletion expressed moderate to high levels of p53 protein in at least 5% of tumor cells. In contrast, p53 protein was undetectable in every tumor that harbored nonsense mutations (frameshift insertion or deletions, stop mutations, and splice junction mutations), a case of strong association (P < .001). However, there were 64 tumors that expressed p53 protein above the 5% level, 43 of which were without a p53 mutation. Among 12 HPV-positive tumors, only four (33%) expressed the p53 protein, compared with 60 cases in the subset of 95 HPV-negative tumors (63%; P = .056). p53 mutation was observed in only one infected tumor (one of 12 HPV-positive cases; 8%) compared with 39 mutations among 95 HPV-positive cases (41%; P = .046). Patients were predominantly men (95 of 107) and smokers (100 of 107). The median age was 59 years (range, 37 to 75 years). The clinical tumor stage was III (34 cases) or IV (73 cases). According to the American Joint Committee on Cancer Staging,35 tumors were classified as T1 and T2 (22 cases) or T3 and T4 (85 cases), and nodal involvement as N0 (47 cases), N1(17 cases), N2 (28 cases), or N3 (15 cases). The primary tumor site was oropharyngeal in 45 cases, laryngeal in 40 cases, and hypopharyngeal in 22 cases. The histologic differentiation was poor in 18 cases and good in 89 cases. Histologic growth pattern was in single cells and thin cellular cords (47 cases) or thick cellular cords and nodules (60 cases). Patient characteristics were not statistically different according to p53 status (P = .11 for tumor stage; P > .20 for other variables; Table 2). There was no significant difference in nodal status whether it was split either in N0 versus N1/N3 or N0/N1 versus N2/N3.
The tumor was measurable in 105 patients (39 mutations). The objective response rate was 63%, and the complete response rate was 22%. Some partial responders have few remaining lesions, most of which actually consist of fibrous scar tissue. These patients with an almost complete response did not differ from complete responders in their response to subsequent radiotherapy and prognosis. Therefore, we thought that it would be useful to analyze these patients together with complete responders as a single subgroup identified as major responders on the basis of a response more than 80% of initial tumor size. The major response rate was 43%. Response data are listed in Table 3.
Univariate analyses of objective response (Table 4) showed that p53 gene status was the only variable significantly associated with objective response, although age, tumor differentiation, and p53 expression were marginally significant (P < .25). Among responders (objective response), the proportion of patients with mutations (17 of 66 cases; 26%) was less than half that observed in nonresponders (22 of 39 cases; 56%; P = .002). p53 gene mutation was associated with a strong reduction in the risk of objective response to chemotherapy (odds ratio, 0.3; 95% confidence interval, 0.1 to 0.6) as compared with wild-type p53. Stepwise logistic-regression analysis showed that p53 gene status was a significant predictor of the risk of objective response to chemotherapy and that age, tumor differentiation, and p53 expression were not. In the multivariate analysis, p53 gene mutation was associated with a 77% reduction in the risk of objective response (adjusted odds ratio, 0.23; 95% confidence interval, 0.10 to 0.57) as compared with wild-type p53. However, p53 expression (adjusted odds ratio, 0.42; 95% confidence interval, 0.17 to 1.06) was retained in the model, and its inclusion was sufficient, together with p53 gene status, to explain much of the variation in the objective response (Hosmer-Lemeshow  2 = 0.014, 2 df; P = .99).
Among major responders, the proportion of patients with mutations (10 of 45 cases; 22%) was less than half the proportion of mutations in the other patients (29 of 60 cases; 48%; P = .006). Tumor stage, node stage, histologic growth pattern, and p53 expression were marginally associated with major response (P < .25). p53 gene mutation was associated with a reduction of 70% in the risk of major response to chemotherapy as compared with wild-type p53 (odds ratio, 0.3; 95% confidence interval, 0.1 to 0.7). Stepwise logistic-regression analysis indicated that p53 gene status (adjusted odds ratio, 0.29; 95% confidence interval, 0.11 to 0.74), p53 expression (adjusted odds ratio, 0.39; 95% confidence interval, 0.16 to 0.98), tumor stage (adjusted odds ratio, 0.31; 95% confidence interval, 0.10 to 0.96) were predictors of the risk of major response. The adequacy of the model was excellent after inclusion of the four selected covariates, including node stage (Hosmer-Lemeshow 2 = 1.249, 8 df; P = .99). p53 gene status was the strongest predictor variable, and its contribution remained significant (P < .02) at all subsequent steps of the analysis, even after tumor stage was entered. Most clinical studies eliminate selectively N2 and N3 disease, which suggests that it would be useful to split the analysis into N0/N1 versus N2/N3. Univariate analysis showed that there was no difference in objective response in patients classified as T3 or T4N0/N1 versus any T N2/N3 (odds ratio, 0.7; 95% confidence interval, 0.3 to 1.5; P = .35). Univariate analysis of major response using N0/N1 versus N2/N3 (odds ratio, 0.5; 95% confidence interval, 0.2 to 1.2; P = .13) did not improve the predictive value of nodal stage compared with the analysis of major response using N0 versus N1 to N3. The multivariate analysis suggested that p53 expression predicted objective or major response to chemotherapy independently from p53 gene status. We asked next whether p53 expression predicted responses to chemotherapy when the analysis was restricted to the subgroup of patients with wild-type p53. Univariate analysis showed that p53 expression had no effect on the risk of objective response in this subgroup, although the risk of major response was significantly reduced by p53 protein expression (odds ratio, 0.3; 95% confidence interval, 0.1 to 1.0; P = .049) as compared with undetectable p53. In a complementary study we asked whether specific mutations would explain the association of overall mutations with poor responses to chemotherapy. Among 39 mutations, 11 (28%) involved the protein domain making contact with DNA, ie, the L3 loop, H2 loop sheet helix, S10 ß strand, and Zinc binding residues.36 Contact and noncontact mutations were equally associated with a reduction in the risk of objective response to chemotherapy (P = .009 and .03, respectively). However, only one of 11 patients (9%) with contact mutation had a major response, as compared with nine of 19 patients (32%) with noncontact mutations and 35 of 66 (53%) with wild-type p53. The difference between the three groups (P = .01) was largely caused by the contrast between the proportions of wild-type p53 and contact mutations (odds ratio, 0.09; 95% confidence interval, 0.01 to 0.73; P = .007). Finally, we asked whether p53 gene status was related to histologic response at the primary tumor site or to nodal response–two major determinants of survival. First, 64 patients (25 mutations) underwent biopsy after chemotherapy. Results for 34 patients (53%) were pathologically negative. The proportion of patients with pathologically negative results among patients with mutations (nine of 25 cases; 36%) was less than that among patients with wild-type p53 (25 of 39 cases; 64%; P = .03). However, pathologic negativity was less strongly predicted by p53 gene status (odds ratio, 0.3; 95% confidence interval, 0.11 to 0.91) than by major response (odds ratio, 0.09; 95% confidence interval, 0.03 to 0.30; P < .001). Second, of 56 patients with a known nodal response, 12 had a complete nodal response. Five of these 12 patients had a mutation, compared with 17 of the 44 patients without a complete nodal response (P = .84).
Our specific purpose was to determine whether p53 gene status as determined by direct sequencing of all coding parts of p53 gene in tumor cells before any treatment in patients with locoregionally advanced SCCHN predicts the initial tumor response to platinum- and fluorouracil-based induction chemotherapy. We found that p53 gene mutations were associated with a three- to four-times reduction in the risk of objective or major responses to chemotherapy. Importantly, patients were treated in our series with a modified Al Sarraf protocol with a reduction in platinum and fluorouracil dosing. We believe that the strong link between p53 gene status and tumor response is a result of the mechanistic involvement of p53 in the apoptotic response to genotoxic drugs. Therefore, we believe that this link would be observed with other dosing or exposure to platinum/fluorouracil chemotherapy, although the corresponding odd ratios can be different. There is currently no other strong predictor of tumor response in patients with locoregionally advanced SCCHN. Tumor stage is weakly predictive of tumor response in the literature.13 Interestingly, although tumor and node stages were not significantly associated with either objective or major response to chemotherapy in univariate analyses, a stepwise logistic-regression model retained these variables and fitted the data well with respect to major response. The association of p53 mutations and tumor stage has been previously noted,37-39 which suggests that p53 mutation rate increases with local progression and, therefore, aggressiveness of disease. We also found a trend for tumors at later stages to contain more frequently p53 mutations than tumors at an earlier stage. Nonetheless, p53 gene status remained clearly predictive of major response in the multivariate analysis when tumor stage was included. p53 expression is often used as a surrogate marker of p53 mutation. However, p53 expression in tumors with mutations depends on the type of mutation,40 as confirmed in our study as well as others.41,42 Although missense mutations induce accumulation of p53 protein, many p53 mutations in squamous cell carcinomas of the respiratory tract are nonsense mutations whose products are truncated proteins,41-43 and these products are undetectable by immunohistochemistry. There is no consensus concerning the levels of p53 protein in tumor cells above which the gene is thought to be overexpressed. Nevertheless, in our study, multivariate analysis showed that p53 expression, when dichotomized with an arbitrarily chosen 5% cutoff, affected both objective and major responses. In addition, low levels of p53 protein in tumor cells were significantly associated with major response in the subgroup of patients with wild-type p53 gene status as determined by sequencing. It is likely that accumulation of p53 in tumor cells reflects inactivation of the p53 protein either by mutation or other mechanisms (probably via mdm2).44 This points to the need to study p53-related molecules to predict chemosensitivity of SCCHN. Such studies might explain outlying observations. Another mechanism by which p53 protein may be inactivated is through binding to HPV gene products.27 This explains why p53 expression or mutation is infrequently found in HPV-positive tumors, as shown in our study as well as others.45 However, the number of HPV-positive cases was too small to draw any conclusion on the relationships of HPV infection on tumor response. A further complexity is that p53 mutations are heterogeneous, ie, some mutations induce a more severe impairment of p53 function than others.46 It was recently shown that specific p53 mutations involving the region of p53 protein making contact with DNA (the L3 loop, H2 loop sheet helix, S10 ß strand, and Zinc binding residues)36 are associated with a poor prognosis in patients with SCCHN.39 We found that DNA contact mutations contributed strongly to the association of overall mutations with poor response to chemotherapy in the analysis of major response. Therefore, specific p53 mutations may predict both a reduced risk to respond to platinum- and fluorouracil-based chemotherapy and a worse prognosis. Although one must express caution regarding any data obtained in the analysis of small subgroups, this suggests that patients with specific p53 mutations may be selected in future trials for other induction regimens, particularly those that include taxanes whose cellular targets are different from classic platinum/fluorouracil regimens.47 Alternatively, patients with specific p53 mutations may be selected for regimens with more exposure to platinum, such as those that include leucovorin.48 Tumor shrinkage at the primary tumor site is the major end point in making the decision for organ preservation. A tumor response to chemotherapy usually predicts a complete response to radiotherapy. Other studies have shown association of p53 mutations with poor responses of SCCHN to radiotherapy25 or esophageal carcinoma to chemoradiotherapy.26 In the study by Koch et al, 25 radiation therapy of SCCHNs with mutated p53 gene was twice as likely to fail locally. This study complements our finding and suggests that patients with p53 mutations are unlikely to benefit from conservative treatment. However, several important reservations should be made clear before concluding that p53 gene status may prove useful in making the decision for an alternative treatment to aggressive surgery. First, given the problems inherent to any retrospective study, we do not show survival data. The survival analysis should be addressed prospectively with homogenous groups of patients with carefully balanced characteristics. Second, the analysis using histologic response and nodal response–two major determinants of survival–gave mixed results. p53 gene status predicted histologic response, albeit less strongly than expected, and did not predict complete nodal response. Clearly, other factors are important in determining the complete disappearance of tumor tissue. The ability to test paraffin-embedded archival tumor certainly is an attractive aspect of our approach. However, the technique remains challenging because the small amount of genetic material in the microdissected tumor samples may cause difficulty in determining p53 gene status. Our sequencing data differ in two main aspects from other series. First, we observed a significant proportion of mutations outside exons 5 to 8, especially in exon 4. This can be easily explained by the a strong bias against mutations in exon 4 in the mutation database because few series have achieved complete sequencing of all coding exons.49 Second, our mutation rate (37%) is in the lower range compared with most other large series (40% to 60%).24,39,41,50 The low rate of mutations in our series may reflect the inclusion of tumors at an early stage and the fact that all samples were taken before any treatment. However, the relatively low proportion of missense mutations in the group of tumors that expressed significant levels of p53 protein makes us suspect that we missed some missense mutations, as did other investigators.24,41,42 Direct sequencing of genomic DNA after enrichment of samples by microdissection is considered as one of the most accurate and sensitive techniques to assess p53 gene status.51 However, recent reports show that alleles that carry point mutations are less efficiently amplified and therefore detected by direct sequencing than alleles with wild-type p53 gene.52 Techniques to enrich samples in alleles that carry point mutations are needed. In conclusion, p53 gene status as determined by direct sequencing of all coding exons predicts the initial tumor shrinkage in response to induction chemotherapy. Because initial tumor response is crucial in making the decision for conservative treatment, our finding warrants further studies. It is likely that p53-related molecules–regulators of p53 function or transcriptional targets of p53–may add to the predictive value of p53 gene status.
Supported by grant no. 1304 from the Association pour la Recherche sur le Cancer, Villejuif, France; grant no. RS-RC/21 from the Comité de Paris de la Ligue Nationale contre le Cancer, Paris, France; and a grant (Bonus Qualité Recherche) from the Conseil Scientifique de la Faculté de Médecine Saint-Antoine, Paris, France. We thank Drs Catherine Fouret and Marie Wislez for helpful discussion; Claire Guéron, Maurice Guéron, and Etienne Roth for reviewing the manuscript; Dr Thierry Soussi for advice; Dominique Dabit for performing HPV detection; Florent Soubrier and Véronique Godard for help in p53 gene sequencing; and Frédéric Commo, Nicole Datti, and Claude Garnier for performing the immunohistochemistry.
1. Vokes EE, Athanasiadis I: Chemotherapy for squamous cell carcinoma of head and neck: The future is now. Ann Oncol 7:15-29, 1996 2. Jacobs C, Goffinet DR, Goffinet L, et al: Chemotherapy as a substitute for surgery in the treatment advanced resectable head and neck cancer: A report from the Northern California Oncology Group. Cancer 60:1178-1183, 1987[Medline] 3. Karp DD, Vaughan CW, Carter R, et al: Larynx preservation using induction chemotherapy plus radiation therapy as an alternative to laryngectomy in advanced head and neck cancer: A long-term follow-up report. Am J Clin Oncol 14:273-279, 1991[Medline] 4. The Department of Veterans Affairs Laryngeal Cancer Study Group: Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer: The Department of Veterans Affairs Laryngeal Cancer Study Group. N Engl J Med 324:1685-1690, 1991[Abstract]
5.
Lefebvre JL, Chevalier D, Luboinski B, et al: Larynx preservation in pyriform sinus cancer: Preliminary results of a European Organization for Research and Treatment of Cancer phase III trial—EORTC Head and Neck Cancer Cooperative Group. Natl Cancer Inst 88:890-899, 1996 6. Kies MS, Gordon LI, Hauck WW, et al: Analysis of complete responders after initial treatment with chemotherapy in head and neck cancer. Otolaryngol Head Neck Surg 93:199-205, 1985[Medline] 7. Forastiere AA: Randomized trials of induction chemotherapy: A critical review. Hematol Oncol Clin North Am 5:725-736, 1991[Medline]
8.
Paccagnella A, Orlando A, Marchiori C, et al: Phase III trial of initial chemotherapy in stage III or IV head and neck cancers: A study by the Gruppo di Studio sui Tumori della Testa e del Collo. J Natl Cancer Inst 86:265-272, 1994
9.
Vokes EE, Weichselbaum RR, Lippman SM, et al: Head and neck cancer. N Engl J Med 328:184-194, 1993 10. Brasnu D, Laccourreye O, Chabardes E, et al: Conservative surgery of the larynx as an alternative to total laryngectomy after neoadjuvant chemotherapy: Initial results. Ann Otolaryngol Chir Cervicofac 110:129-133, 1993[Medline] 11. Bailet JW, Sercarz JA, Abemayor E, et al: The use of positron emission tomography for early detection of recurrent head and neck squamous cell carcinoma in postradiotherapy patients. Laryngoscope 105:135-139, 1995[Medline] 12. Sassler AM, Esclamado RM, Wolf GT: Surgery after organ preservation therapy: Analysis of wound complications. Surg 121:162-165, 1995 13. Wolf GT, Makuch RW, Baker SR: Predictive factors for tumor response to preoperative chemotherapy in patients with head and neck squamous carcinoma: The Head and Neck Contracts Program. Cancer 54:2869-2877, 1984[Medline]
14.
Spaulding MB, Fischer SG, Wolf GT: Tumor response, toxicity, and survival after neoadjuvant organ-preserving chemotherapy for advanced laryngeal carcinoma: The Department of Veterans Affairs Cooperative Laryngeal Cancer Study Group. J Clin Oncol 12:1592-1599, 1994
15.
Johnston PG, Mick R, Recant W, et al: Thymidylate synthase expression and response to neoadjuvant chemotherapy in patients with advanced head and neck cancer. J Natl Cancer Inst 89:308-313, 1997
16.
Cheradame S, Etienne MC, Formento P, et al: Tumoral-reduced folates and clinical resistance to fluorouracil-based treatment in head and neck cancer patients. J Clin Oncol 15:2604-2610, 1997
17.
Reynolds T: Research on drug resistance unearths many molecules, many mechanisms. J Natl Cancer Inst 90:1120-1122, 1998
18.
Evan G, Littlewood T: A matter of life and cell death. Science 281:1317-1322, 1998
19.
de Stanchina E, McCurrach ME, Zindy F, et al: E1A signaling to p53 involves the p19(ARF) tumor suppressor. Genes Dev 12:2434-2442, 1998
20.
Zindy F, Eischen CM, Randle DH, et al: Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev 12:2424-2433, 1998 21. Lowe SW, Ruley HE, Jacks T, et al: p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 74:957-967, 1993[Medline]
22.
Liu TJ, el-Naggar AK, McDonnell TJ, et al: Apoptosis induction mediated by wild-type p53 adenoviral gene transfer in squamous cell carcinoma of the head and neck. Cancer Res 55:3117-3122, 1995 23. Pirollo KF, Hao Z, Rait A, et al: p53 mediated sensitization of squamous cell carcinoma of the head and neck to radiotherapy. Oncogene 14:1735-1746, 1997[Medline] 24. Clayman GL, el-Naggar AK, Lippman SM, et al: Adenovirus-mediated p53 gene transfer in patients with advanced recurrent head and neck squamous cell carcinoma. J Clin Oncol 16:2221-2232, 1998[Abstract]
25.
Koch WM, Brennan JA, Zahurak M, et al: p53 mutation and locoregional treatment failure in head and neck squamous cell carcinoma. J Natl Cancer Inst 88:1580-1586, 1996 26. Ribeiro U Jr, Finkelstein SD, Safatle-Ribeiro AV, et al: p53 sequence analysis predicts treatment response and outcome of patients with esophageal carcinoma. Cancer 83:7-18, 1998[Medline]
27.
Scheffner M, Huibregtse JM, Howley PM: Identification of a human ubiquitin-conjugating enzyme that mediates the E6-AP-dependent ubiquitination of p53. Proc Natl Acad Sci U S A 91:8797-8801, 1994
28.
Pfister DG, Harrison LB, Strong EW, et al: Organ-function preservation in advanced oropharynx cancer: Results with induction chemotherapy and radiation. J Clin Oncol 13:671-680, 1995 29. Hsu IC, Metcalf RA, Sun T, et al: Mutational hotspot in the p53 gene in human hepatocellular carcinomas. Nature 350:427-428, 1991[Medline] 30. Lazarus P, Garewal HS, Sciubba J, et al: A low incidence of p53 mutations in pre-malignant lesions of the oral cavity from non-tobacco users. Int J Cancer 60:458-463, 1995[Medline]
31.
Cariello NF, Douglas GR, Gorelick NJ, et al: Databases and software for the analysis of mutations in the human p53 gene, human hprt gene and both the lacI and lacZ gene in transgenic rodents. Nucleic Acids Res 26:200-202, 1998
32.
Resnick RM, Cornelissen MT, Wright DK, et al: Detection and typing of human papillomavirus in archival cervical cancer specimens by DNA amplification with consensus primers. J Natl Cancer Inst 82:1477-1484, 1990 33. Anwar K, Nakakuki K, Imai H, et al: Over-expression of p53 protein in human laryngeal carcinoma. Int J Cancer 53:952-956, 1993[Medline] 34. Fouret P, Dabit D, Sibony M, et al: Expression of p53 protein related to the presence of human papillomavirus infection in precancer lesions of the larynx. Am J Pathol 146:599-604, 1995[Abstract] 35. Beahrs OH, Henson DE, Hutter RVP, et al (eds): American Joint Commitee on Staging: Manual for Staging of Cancer. Philadelphia, PA, Lippincott, 1992
36.
Cho Y, Gorina S, Jeffrey PD, et al: Crystal structure of a p53 tumor suppressor-DNA complex: Understanding tumorigenic mutations. Science 265:346-355, 1994
37.
Boyle JO, Hakim J, Koch W, et al: The incidence of p53 mutations increases with progression of head and neck cancer. Cancer Res 53:4477-4480, 1993 38. Lee NK, Ye YW, Chen J, et al: p53, retinoblastoma, and human papillomavirus in squamous cell carcinoma and adjacent normal mucosa of the upper aerodigestive tract. Head Neck Surg 119:1125-1131, 1993 39. Erber R, Conradt C, Homann N, et al: TP53 DNA contact mutations are selectively associated with allelic loss and have a strong clinical impact in head and neck cancer. Oncogene 16:1671-1679, 1998[Medline] 40. Bodner SM, Minna JD, Jensen SM, et al: Expression of mutant p53 proteins in lung cancer correlates with the class of p53 gene mutation. Oncogene 7:743-749, 1992[Medline]
41.
Brennan JA, Boyle JO, Koch WM, et al: Association between cigarette smoking and mutation of the p53 gene in squamous-cell carcinoma of the head and neck. N Engl J Med 332:712-717, 1995 42. Melhem MF, Law JC, el-Ashmawy L, et al: Assessment of sensitivity and specificity of immunohistochemical staining of p53 in lung and head and neck cancers. Am J Pathol 146:1170-1177, 1995[Abstract]
43.
Wang YC, Chen CY, Chen SK, et al: High frequency of deletion mutations in p53 gene from squamous cell lung cancer patients in Taiwan. Cancer Res 58:328-333, 1998 44. Midgley CA, Lane DP: p53 protein stability in tumour cells is not determined by mutation but is dependent on Mdm2 binding. Oncogene 15:1179-1189, 1997[Medline] 45. Portugal LG, Goldenberg JD, Wenig BL, et al: Human papillomavirus expression and p53 gene mutations in squamous cell carcinoma. Otolaryngol Head Neck Surg 123:1230-1234, 1997 46. Ory K, Legros Y, Auguin C, et al: Analysis of the most representative tumour-derived p53 mutants reveals that changes in protein conformation are not correlated with loss of transactivation or inhibition of cell proliferation. EMBO J 13:3496-3504, 1994[Medline] 47. Forastiere AA, Shank D, Neuberg D, et al: Final report of a phase II evaluation of paclitaxel in patients with advanced squamous cell carcinoma of the head and neck: An Eastern Cooperative Oncology Group trial (PA390). Cancer 82:2270-2274, 1998[Medline] 48. Kies MS, Haraf DJ, Athanasiadis I, et al: Induction chemotherapy followed by concurrent chemoradiation for advanced head and neck cancer: Improved disease control and survival. J Clin Oncol 16:2715-2721, 1998[Abstract] 49. Casey G, Lopez ME, Ramos JC, et al: DNA sequence analysis of exons 2 through 11 and immunohistochemical staining are required to detect all known p53 alterations in human malignancies. Oncogene 13:1971-1981, 1996[Medline] 50. Ahomadegbe JC, Barrois M, Fogel S, et al: High incidence of p53 alterations (mutation, deletion, overexpression) in head and neck primary tumors and metastases: Absence of correlation with clinical outcome–Frequent protein overexpression in normal epithelium and in early non-invasive lesions. Oncogene 10:1217-1227, 1995[Medline] 51. Hartmann A, Blaszyk H, Kovach JS, et al: The molecular epidemiology of p53 gene mutations in human breast cancer. Trends Genet 13:27-33, 1997[Medline] 52. Barnard R, Futo V, Pecheniuk N, et al: PCR bias toward the wild-type k-ras and p53 sequences: Implications for PCR detection of mutations and cancer diagnosis. Biotechniques 25:684-691, 1998[Medline] Submitted December 22, 1998; accepted July 29, 1999. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
