p53 gene mutations are associated with poor survival in low and low-intermediate risk diffuse large B-cell lymphomas

K. Leroy1,+, C. Haioun2,§, E. Lepage3, N. Le Métayer1, F. Berger4, E. Labouyrie5, V. Meignin6, B. Petit7, C. Bastard8, G. Salles9, C. Gisselbrecht10, F. Reyes2 and Ph. Gaulard1

1 Département de Pathologie, 2 Service d’Hématologie and 3 Unité d’Informatique Médicale, Hôpital Henri Mondor, AP-HP, Créteil; 4 Service d’Anatomie Pathologique and 9 Service d’Hématologie, Centre Hospitalier Lyon-Sud, Pierre-Bénite; 5 Service d’Anatomie Pathologique, Hôpitaux de Brabois, CHU Nancy; 6 Service d’Anatomie Pathologique and 10 Institut d’Hématologie, Hôpital Saint-Louis, AP-HP, Paris; 7 Service d’Anatomie Pathologique, Hôpital Dupuytren, Limoges; 8 Laboratoire de Génétique Oncologique, Centre Henri Becquerel, Rouen, France

Received 9 November 2001; revised 21 January 2002; accepted 12 February 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

p53 alterations have been associated with a poor prognosis in aggressive B-cell lymphoma. We investigated the clinical relevance of p53 status in diffuse large B-cell lymphoma (DLBCL), focusing on patients who belong to lower risk groups of the international prognostic index and were uniformly treated. We aimed to determine whether this biological marker could identify among such patients those with a pejorative outcome who could benefit from a distinct therapeutic approach.

Patients and methods:

We studied 69 patients presenting with no, one (low-risk, n = 40) or two (low-intermediate risk, n = 29) risk factors treated with an anthracyclin-containing induction regimen. p53 exons 5–8 mutations were screened for using denaturing gradient gel electrophoresis and confirmed by direct sequencing. Immunohistochemical detection of p53 protein and of its downstream target p21 were also evaluated in 60 of 69 cases.

Results:

p53 mutations were detected in 16 of 69 (23%) lymphoma samples. The presence of a p53 gene mutation affected survival (P = 0.01), with a 6-year survival rate estimated to be 44% in mutated patients, compared with 79% in non-mutated ones. Using a stepwise Cox model, p53 mutation constituted the only parameter affecting survival (relative risk = 2.7, P = 0.03). A p53+/p21 immunohistochemical pattern (n = 15), suggestive of a disrupted p53 function, strongly correlated with p53 gene status and was associated with a lower 6-year survival rate when compared with a p53 or p53+/p21+ phenotype (47% versus 74%, P = 0.05).

Conclusions:

p53 alterations constitute a pejorative biological indicator able to discriminate among clinically defined lower risk patients with DLBCL.

Key words: diffuse large B-cell lymphomas, p53, survival


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Diffuse large B-cell lymphomas (DLBCL) represent 30% to 40% of adult non-Hodgkin’s lymphomas (NHL) [1]. This lymphoma entity, as defined in the Revised European–American Lymphoma (REAL) [1] and WHO classifications [2], has been recognized from the start as relatively heterogeneous regarding histological subtypes, genetic alterations and clinical evolution. At the present time, patient outcome prediction and treatment strategies rely on the International Pronostic Index (IPI) [3], which identifies four different patient risk subgroups. This predictive model is based on clinical characteristics that are supposed to reflect the tumor’s growth and invasive potential, and the patient’s response to the tumor and ability to tolerate intensive therapy [4]. However, IPI subgroups remain heterogeneous and molecular markers could help to identify within these subgroups those patients who will do well, and those who will relapse, so that treatment strategies can be better tailored. Indeed, several studies have correlated the expression levels of anti-apoptotic proteins such as Bcl2 [5] or survivin [6], or adhesive proteins such as CD44 [7], to clinical course and shown that these factors add substantial prognostic information to the IPI score.

The p53 gene encodes a 393 amino acid phosphoprotein that regulates the transcription of several genes [8]. It is now considered that the main function of the p53 protein is to integrate cellular responses to stress [9]. In response to various internal or external signals, such as hypoxia, DNA damage or oncogenic stimulus, the wild-type p53 protein undergoes post-translational modifications, accumulates within the nucleus of the cells and induces cell cycle arrest or apoptosis [10]. Alterations of the p53 tumor suppressor gene may be considered as a model for carcinogenesis, since this gene is mutated and/or deleted in >50% of human cancers [9], its germinal mutations predispose to neoplasia [11], its somatic mutation pattern can be used as an epidemiological marker of environmental mutagens [12], and last but not least, its protein product regulates both cell growth and cell death [9]. In addition, several studies have shown that wild-type p53 function was required for apoptosis induced by genotoxic damage [13, 14], suggesting that p53 functional status could be correlated with treatment response. Despite a large amount of data in the field, the clinical relevance of p53 alterations is still a matter for debate in the case of solid tumors [15], holding true for some cancers [16] but not others [17, 18]. As for lymphoid proliferations, p53 gene mutations are associated with the histological transformation of follicular lymphoma [19, 20] and B-cell chronic lymphocytic leukemia [21], and with blastic variants of mantle cell lymphoma [22]. In B-cell chronic lymphocytic leukemia, as well as in multiple myeloma, p53 deletions have been associated with shorter survival and poor response to chemotherapy [23, 24]. A few studies have reported that p53 mutations are associated with drug resistance, shorter progression-free survival and decreased overall survival in aggressive B-cell lymphomas [2527]. However, since these studies were conducted on patients who were heterogeneous with regard to histological subtypes, clinical risk factors and treatment regimens, the clinical impact of p53 mutations remains unclear.

This prompted us to investigate the clinical relevance and the predictive value of p53 gene mutations in a series of 69 patients with DLBCL, a low or low-intermediate risk of disease according to the IPI and who were treated uniformly. Indeed, we hypothesized that among these lower risk patients, those with a p53 mutation would benefit from more intensive therapies and that higher risk group patients might have accumulated numerous genetic changes and confounding adverse clinical factors, which could obscure the impact of p53 on outcome. All the patients studied were enrolled in the LNH87 and LNH93 trials of the Groupe d’Etude des Lymphomes de l’Adulte. p53 mutations in exons 5–8 were screened for using denaturing gradient gel electrophoresis (DGGE) and confirmed by direct sequencing. In addition, some authors suggested that the combined immunohistochemical analysis of p53 and one of its downstream target p21/WAF1 [8]—a cyclin-dependent kinase inhibitor—could be a valuable means to assess p53 functional status in lymphomas [28, 29]. These surrogate markers, which could readily be used in routine laboratory practice, were evaluated in 60 of these patients and correlated with p53 mutations and outcome.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patient selection
Sixty-nine patients were studied. They all fulfilled the following criteria: (i) diagnosis of de novo DLBCL confirmed by pathological review; (ii) low or low-intermediate risk disease according to the IPI; (iii) induction treatment by four courses of ACVBP or NCVBP (adriamycine or mitoxantrone, cyclophosphamide, vindesine, bleomycin, prednisone) followed, in cases of good response, by the LNH84 sequential consolidative treatment [30]; and (iv) frozen tumor sample or tumoral DNA available for analysis within six institutions in France (Créteil, Limoges, Lyon, Nancy, Saint-Louis, Rouen). These patients were treated in the LNH87 and LNH93 protocols between December 1987 and October 1997. Their main clinical characteristics were similar to those of the group of 1048 patients with low or low-intermediate risk DLBCL involved in the same protocols during the same period, apart from age (age >60 years, 10% versus 20%; P = 0.05) and extranodal extension [extranodal sites (ENS) >1, 10% versus 20%; P = 0.04]. The repartition of the basis of IPI was similar between the two groups (low, 58% versus 65%; low-intermediate, 42% versus 35%; P = 0.22). Complete remission was achieved in 81% and 76% of the patients, respectively (P = 0.37). Disease dissemination was evaluated before treatment by physical examination, bone marrow biopsy and computed tomography (CT) scan of the chest and abdomen. Patients were staged according to the Ann Arbor system. The number of ENS and the largest tumor mass diameter were also determined. Performance status (PS) was based on the Eastern Cooperative Oncology Group (ECOG) scale. The serum lactate dehydrogenase (LDH) level was expressed as the ratio over the maximal normal value. Response to therapy was evaluated after induction treatment and complete remission (CR) was defined as the disappearance of all clinical evidence of disease and normalization of all laboratory values, radiographs, CT scans and bone marrow histology.

DNA samples
High molecular weight DNA was prepared from frozen tumoral cryostat sections (60 cases), and from tumoral cell pellets (nine cases) using proteinase K digestion, phenol–chloroform extraction and ethanol precipitation, as described previously [31]. An histological or cytological control was performed in each case.

Denaturing gradient gel electrophoresis
DGGE screening for p53 mutations located in exons 5–8 was performed essentially as described previously [32]. Five PCRs are required to explore the presence of mutations in these exons. We used the primer pairs described by Hamelin et al. [32], with a GC clamp associated with the antisense primer for DGGE analysis of exon 8. DNAs were amplified in a Perkin Elmer 2400 thermocycler (Applied Biosystems, Foster City, CA, USA) in 25 µl reaction mixture containing 200 ng of tumoral genomic DNA, 0.3 µM of each primer, 200 µM of each deoxynucleotide and 0.5 U Taq Gold DNA polymerase (Applied Biosystems) in GeneAmp PCR buffer II (Tris 10 mM pH 8.3, KCl 50 mM) with MgCl2 at a final concentration of 1.2 mM (PCR exon 8) or 2 mM (PCR exon 5p, 5d, 5+6 and 7). The reactions were heated at 95°C for 10 min, amplified by 38 cycles of 94°C for 30 s, 55°C (PCR exon 5p, 8) or 62°C (PCR exon 5d, 5+6, 7) for 1 min and 72°C for 90 s, followed by a final 7 min elongation step. Heteroduplex formation was enhanced by: (i) a round of denaturation (98°C for 10 min) and hybridization (55°C or 62°C for 30 min) at the end of the PCR program; and (ii) by adding 50 ng of normal liver genomic DNA to each reaction. In our experience, this avoids misinterpretation due to the fact that mutated homoduplexes gel migration is sometimes very close to, and undistinguishable from, that of normal homoduplexes, without significantly lowering the sensitivity of this method. In this study, PCR amplifications were performed both with and without the addition of normal liver genomic DNA in 20 cases, with similar results.

Samples were loaded onto 16 x 22 cm, 0.75-mm thick, 6.5% polyacrylamide gels (acrylamide–bisacrylamide 29:1) containing a denaturant gradient (100% denaturant; 7 M urea, 40% v/v deionized formamide) parallel to the direction of electrophoresis. Gels were run in a DGGE system (C.B.S., Del Mar, CA, USA) in 1x TAE buffer (40 mM Tris acetate, 20 mM sodium acetate, 1 mM EDTA pH 7.4) maintained at 60°C. Electrophoresis was performed in a 50% to 100% gradient for 15 h (exon 5p) or for 4 h (exon 5d), a 30% to 100% gradient for 6 h (exons 5+6), a 30% to 80% gradient for 5 h (exon 7) or a 20% to 70% gradient for 7 h (exon 8). After electrophoresis, gels were stained with ethidium bromide and photographed using a UV transilluminator. Control DNAs (from human leukemic cell lines HUT78 and CEM, and colon tumors) exhibiting known mutations in the various PCR fragments explored were systematically amplified and run in parallel with the lymphoma samples studied.

Sequencing
Tumoral DNAs exhibiting an abnormal profile upon DGGE analysis were PCR amplified with p53 primer pairs containing M13 reverse or –21M13 sequences at their 5' end. Amplicons used for sequencing corresponded to fragments spanning nucleotides 13 004–13 457, 13 947–14 140 and 14 403–14 804 of p53 human gene locus (GenBank locus name HSP53G, GI: 35214). PCR products were run in 1.5% agarose gels containing ethidium bromide to ascertain both the amount and the quality of PCR products obtained, diluted 1:5 to 1:10 in water and used for direct sequencing on both strands with Dye-Primer Ready-Reaction mixes (Applied Biosystems). Excising the abnormal DGGE fragments from the acrylamide gel, followed by re-amplification and direct sequencing, was required for one sample (R1397) harboring a codon 248 missense mutation that was thus qualified as ‘minor’. The intron 6 mutations were identified upon direct sequencing of the P8-Pa9 PCR products using a Dye-Terminator Ready-Reaction mix, and the 5'-TCAGCATCTTATCCGAGTGG-3' oligonucleotide located in p53 exon 6, 80 nucleotides upstream from the splice site. Sequencing reactions were loaded on a 4.25% acrylamide–7 M urea denaturing gel using a ABI 377 (Applied Biosystems) sequencing apparatus, and mutation analysis performed with Sequence navigator software (Applied Biosystems).

Immunohistochemical detection of p53 and p21 proteins
Sixty cases were studied for p53 expression by immunohistochemistry (IHC), performed on deparaffinized tissue sections using an indirect immunoperoxidase method with a Nexes automat (Ventana Medical Systems, Tucson, AZ, USA) according to the manufacturer’s instructions. p21 expression was also evaluated in the cases showing p53 overexpression. The slides were pre-treated using microwave oven heating (three cycles of 5 min in 0.01 M citrate buffer, pH 7.3), and incubated for 30 min at 37°C with monoclonal antibodies directed against p53 protein (DO-7; Dako SA, Glostrup, Denmark; dilution 1:20) or p21/WAF1 protein (EA-10; Oncogene Sciences, Uniondale, NY, USA; dilution 1:50). Positive control slides were included in all tests. Each slide was examined by two authors (K.L., P.G.) without any knowledge of the clinical and molecular data. A semi-quantitative evaluation of p53 and p21 protein expression was performed by estimating the percentage of positive tumor cells: –, 0% to 10%; +, 10% to 25%; ++, 25% to 50%; and +++, 50% to 100%, with a cut-off point of 10% positive cells as reported by several authors [28, 29].

Statistical analysis
Patient characteristics and CR rates were compared using the {chi}2 test. Overall survival time was calculated from the date of diagnosis until death, last follow-up examination or stopping date, 1 December 1999. Survival curves were estimated using the Kaplan–Meier product-limit method [33] and were compared using the log-rank test. Univariate analyses were made using the {chi}2 test and the log-rank test. A multivariate regression analysis according to Cox’s proportional hazards regression model [34], with overall survival as the dependent variable, was used to adjust the effect of p53 mutations, as well as p53 expression and {Delta}p53 (p53+/p21) immunophenotype for potential independent prognostic factors. A P value of <0.05 (two-sided test) was considered to indicate statistical significance. All calculations were performed with SAS software, version 6.10 (SAS Institute, Cary, NC, USA).


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
p53 gene mutations
Among the 69 lymphoma samples studied, 18 exhibited an abnormal DGGE profile (Figure 1). Two cases presented a DNA polymorphism of codon 213 (Arg CGA->CGG), observed at a low frequency (3% to 11%) in the European population [35, 36]. In the remaining 16 cases, sequence analysis revealed missense mutations in exon 5 (six cases), exon 7 (four cases) or exon 8 (three cases); a nonsense mutation in exon 8 (one case); a 9-nucleotide deletion in exon 6 (one case); and intronic point mutations (two cases) (Figure 2 and Table 1). These intronic mutations do not constitute known p53 gene polymorphisms [37], and have not been observed in 120 non-lymphoid tumors explored in our laboratory in the same DGGE conditions. Thus, although these mutational events do not constitute classical mutational hotspots, and in the absence of matching non-tumoral DNA that could rule out DNA polymorphisms, both samples were considered as mutated. Besides, one of these two lymphoma samples (R1397) also harbored a missense mutation in exon 7.



View larger version (52K):
[in this window]
[in a new window]
 
Figure 1. Detection of p53 exon 5 mutations by DGGE. DGGE profile observed after amplification of normal DNA (lane 1), and DLBCL DNAs presenting mutations of p53 codon 135 (lane 2), 157 (lane 3), 173 (lane 4), 175 (lane 5) and 179 (lane 6).

 


View larger version (27K):
[in this window]
[in a new window]
 
Figure 2. Identification of p53 mutations by direct sequencing. Electrophoregrams observed upon fluorescent sequencing of DLBCLs PCR products. (A) Codon 157 GTC->GGC; (B) codon 173 GTG->TTG; (C) codon 175 CGC->CAC. The arrow indicates the mutated nucleotide, and the codon affected by the mutation is underlined.

 

View this table:
[in this window]
[in a new window]
 
Table 1.  p53 mutations identified in 16 DLBCLs, p53 protein expression and clinical outcome
 
Correlation of p53 mutations with clinical characteristics and outcome
Table 2 shows the characteristics of the 69 patients, according to the presence or absence of p53 mutations. The two groups did not differ significantly in their main clinical characteristics apart from the fact that patients with a p53 mutated lymphoma presented with a bulky disease more frequently. There was no significant difference in the proportion of patients who achieved CR (83% versus 69%; P = 0.16). However, with a median follow-up of the surviving patients of 80 months (range 16–141), patients with a p53 mutated lymphoma exhibited a lower 6-year survival rate than those with wild-type p53: 44% [confidence interval (CI) 19–68] versus 79% (CI 68–90) (P = 0.01; Figure 3). Univariate analysis showed that elevated LDH levels were also significantly associated with decreased overall survival (P = 0.05), whereas age, PS, stage and number of extranodal sites did not show significant prognostic value. A stepwise Cox model incorporating prognostic factors from the IPI (i.e. age, PS, stage, LDH level and number of extranodal sites) and the p53 gene status identified the presence of a p53 gene mutation as the only parameter significantly correlated with a decreased survival [relative risk (RR) = 2.7; P = 0.03]. In order to explore the impact of p53 mutation on survival among the two risk groups defined on the IPI basis, we performed an additional Cox model incorporating the p53 mutation status and the risk group (low/low-intermediate). This analysis showed that the prognostic significance of p53 mutation on survival was independent of the risk group (P = 0.02). In addition, 63% of p53 mutated patients experienced either progression, relapse or lymphoma-related death, compared with 25% for non-mutated ones, during the first 2 years (Figure 4).


View this table:
[in this window]
[in a new window]
 
Table 2.  Clinical characteristics of the 69 DLBCL patients according to p53 gene status
 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 3. Kaplan–Meier curves of overall survival of 69 patients with DLBCL according to p53 gene status. Cases with a p53 mutated lymphoma, n = 16; cases with a wild-type p53 lymphoma, n = 53.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 4. Kaplan–Meier curves of event free survival of 69 patients with DLBCL according to p53 gene status. Cases with a p53 mutated lymphoma, n = 16; cases with a wild-type p53 lymphoma, n = 53.

 
p53 IHC expression: correlations with p53 mutations and clinical outcome
Among the 60 lymphoma samples studied for p53 expression, 16 (27%) displayed p53 overexpression, as defined by p53 nu-clear staining in at least 10% of the neoplastic cells (Figure 5). p53 protein expression appeared to be strongly correlated with p53 gene status (P = 0.001), since most of the p53 mutated samples (12 of 14, 85%) exhibited nuclear staining (Table 1), whereas non mutated tumors were essentially p53 negative (42 of 46 samples, 91%). In addition, one of the two p53 mutated, IHC-negative samples (R1415) had a nonsense truncating mutation, which can result in an unstable transcript and/or protein, and thus no p53 accumulation. p53 IHC-positive patients demonstrated a trend towards a shorter 6-year survival rate compared with those without p53 overexpression, that did not reach statistical significance (74% versus 50%; P = 0.09).



View larger version (125K):
[in this window]
[in a new window]
 
Figure 5. Immunohistochemical detection of p53 protein in DLBCLs. Representative cases showing (A) strong nuclear labeling of virtually all tumor cells; (B) nuclear staining in 10% to 25% of tumor cells; and (C) negative for p53 expression.

 
Definition of a {Delta}p53 immunophenotype and correlation with clinical outcome
p53 protein immunohistochemical detection reflects either a p53 gene mutation affecting p53 turnover, or wild-type p53 accumulation induced by an endogeneous stress [9]. In NHLs, some authors have suggested that combined IHC analysis of p53 and its downstream target p21 enables the recognition of a p53+/p21 immunophenotype, frequently associated with p53 gene mutations, whereas a p53+/p21+ phenotype merely reflects the activation of wild-type p53 [28, 29]. These studies prompted us to investigate p21 protein expression in the 16 p53-positive samples. p21 expression was observed in >10% of tumoral cells in only one sample, which was normal upon DGGE analysis, and p21 was scored as negative in all the other cases, comprising 12 p53 mutated cases and three cases with normal p53 exons 5–8 DGGE profiles (Table 3). Definition of a {Delta}p53 phenotype (p53+/p21) as a surrogate for a disrupted p53 function appeared to be highly correlated with p53 gene status, since 55 of 60 (91%) cases were concordant for p53 DGGE analysis and p53/p21 IHC analysis. In addition, this {Delta}p53 phenotype showed clinical relevance, since {Delta}p53 patients (n = 15) disclosed a significantly lower 6-year survival rate compared with patients with a p53 or a p53+/p21+ immunophenotype (n = 45) (74% versus 47%; P = 0.05) (Figure 6). However, using a multivariate Cox analysis, the independent prognostic value of {Delta}p53 in predicting survival could not be established (P = 0.07). Interestingly, p53 mutations remained an independent prognostic factor when the analysis was limited to the 60 patients studied for p53 and p21 protein expression (P = 0.02).


View this table:
[in this window]
[in a new window]
 
Table 3.  Distribution of the immunohistochemical pattern of p53 and p21 expression in relation with p53 gene status
 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 6. Kaplan–Meier curves of overall survival of 60 patients with DLBCL according to {Delta}p53 immunophenotype. Cases with a {Delta}p53 (p53+/p21) immunophenotype, n = 15; compared with p53 or p53+/p21+ cases, n = 45.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
We have investigated the relationship between p53 gene mutations, clinical characteristics and outcome in a series of patients with de novo DLBCL, a low or low-intermediate risk of disease, who were treated uniformly. In 1997, Ichikawa et al. [25] reported the poor prognostic value of p53 gene mutations in a series of 102 patients with aggressive B-cell lymphoma, including 59 DLBCL patients. When these patients were retrospectively categorized according to the IPI, it appeared that this poor prognostic value was limited to the subgroup of low and low-intermediate risk patients. In the present study, we demonstrate that p53 mutations are significantly associated with a decreased overall survival and thus constitute a poor prognostic factor in such lower risk patients .

The frequency of p53 gene mutations that was observed in this series of patients (23%), their type—i.e. mostly missense mutations—and their location, are overall in agreement with previous reports describing the p53 mutation pattern in this disease [25, 26, 28, 29]. Obviously, we cannot rule out the possibility of mutations outside the exons 5–8, which were not explored in this study, especially in the three cases presenting an IHC p53+/p21 phenotype. However, >90% of p53 mutations reported so far affect the central core domain involved in DNA binding, and mutations outside exons 5–8 appear to be rare [12]. Two lymphoma samples presented the same unusual mutation, located in intron 6. One of these two samples also displayed a minor missense mutation of codon 248, and both exhibited p53 protein accumulation. The deleterious effect associated with such intronic alterations cannot be formally assessed. However, nucleotide alterations in p53 intron 6 have been reported to be associated with cancer predisposition syndromes, p53 protein accumulation and decreased in vitro chemotherapy-induced apoptosis [3840], suggesting that intron 6 mutations might somehow dysregulate p53 function and promote tumor formation.

We were also interested to explore the relationship between the location of the different mutations identified and the clinical course. Quite interestingly, it appeared that mutations occurring in the p53 protein domains in close contact with DNA [41] (i.e. those located in loop L3, sheet S10 and helix H2) were associated with a dramatic clinical course. Indeed, six of seven patients harboring such mutations (see Table 1) died within 2 years, whereas only three of nine patients with a p53 mutation located outside these regions died, which represents a death rate similar to that observed in non-mutated patients (12 deaths among 43 patients). The limited number of mutated cases precludes a statistical analysis of the prognostic load associated with p53 mutation location in this series of patients. However, a poor prognosis and/or poor chemotherapy response specifically associated with L3 domain mutations have already been reported in the case of colorectal [42] and breast cancers [43].

In this series, we observed that p53 gene mutations were more frequent in patients presenting with a bulky disease. This association was not reported in previous studies concerning p53 alterations in B-cell lymphomas, but it might have been obscured by the heterogeneity of the lymphomas studied with regard to histological types and clinical stage. The higher frequency of p53 gene mutations in bulky disease might either reflect an increased risk of secondary genetic events parallel to tumoral growth or be related to an increased resistance to hypoxia of p53 mutated cells [44], as it has been suggested in the case of solid tumors [45].

In contrast to previous studies [25, 27], we failed to demonstrate a significant relationship between the presence of a p53 mutation and response to chemotherapy in this series of patients, but it should be stressed that the response was evaluated at the end of the induction regimen in this study, and much later (after chemotherapy completion) in the previous reports. Contrary to what could be expected from in vitro studies demonstrating cell killing resistance associated with p53 mutations [13, 46], it might be that p53 alterations do not, per se, prevent the initial tumoral burden reduction induced by chemotherapy. Since patients with a p53-mutated lymphoma frequently experienced short-term relapse, one might hypothesize that p53 mutations favor the rapid emergence of resistant clones, as a consequence of the increased genomic instability that has been linked to p53 alterations [47, 48].

The evaluation of p53 status with molecular techniques cannot easily be carried out in routine laboratory practice. Several investigators have attempted to use the IHC analysis of p53 protein accumulation or both p53 and p21 proteins, as a surrogate for p53 mutations [28, 29]. This combined evaluation, which relies on the definition of a {Delta}p53 phenotype (p53+/p21), was recently reported to be a predictor of treatment failure and poor prognosis in a B- and T-cell NHL population-based study [49]. In our series of patients, p53 IHC analysis appeared to be slightly less efficient than the combined p53 and p21 analysis in discriminating between p53 mutated and non-mutated lymphomas, and use of the {Delta}p53 immunophenotype did show clinical relevance with respect to overall survival. However, considering that p53 and p21 IHC staining may vary between laboratories, the definition and validation of optimal cut-off values are needed to routinely apply these techniques.

Altogether, this study allowed us to assess the poor prognosis associated with p53 mutations in a homogenous series of patients with low and low-intermediate risk DLBCL. The clinical use of this prognostic marker should be useful to identify those among these patients who require distinct therapeutic approaches. Moreover, further studies would be helpful to evaluate the prognostic significance of the genomic location of the mutation and confirm the clinical relevance of the {Delta}p53 immunophenotype.


    Acknowledgements
 
The authors are indebted to Antoine Allain, Nicolas Nio and Fabrice Jehanno for expert assistance with data management. The authors thank the following clinicians and pathologists who participated in this study: D. Bordessoule, J. Brière, J. Diebold, C. Duval, P. Lederlin, T. Molina, P. Morel, H. Tilly and J. M. Vignaud. This study was supported in part by grants from the French Ministry of Health (Programme Hospitalier de Recherche Clinique).


    Footnotes
 
+ Correspondence to: Dr K. Leroy, Département de Pathologie, Hôpital Henri Mondor, 51 avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France. Tel: +33-1-49-81-27-42; Fax: +33-1-49-81-27-33; E-mail: karen.leroy@hmn.ap-hop-paris.fr Back

§ K.Leroy and C.Haioun contributed equally to this work. Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Harris NL, Jaffe ES, Stein H et al. A revised European–American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994; 84: 1361–1392.[Free Full Text]

2. Harris NL, Jaffe ES, Diebold J et al. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting—Airlie House. J Clin Oncol 1999; 17: 3835–3849.[Abstract/Free Full Text]

3. Shipp MA, Harrington DP, Anderson JR et al. A predictive model for aggressive non-Hodgkin’s lymphoma. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993; 329: 987–994.[Abstract/Free Full Text]

4. Shipp MA. Prognostic factors in aggressive non-Hodgkin’s lymphoma: who has ‘high-risk’ disease? Blood 1994; 83: 1165–1173.[Abstract/Free Full Text]

5. Hermine O, Haioun C, Lepage E et al. Prognostic significance of bcl2 protein expression in agressive non-Hodgkin’s lymphoma. Blood 1996; 87: 265–272.[Abstract/Free Full Text]

6. Adida C, Haioun C, Gaulard P et al. Prognostic significance of survivin expression in diffuse large B-cell lymphomas. Blood 2000; 96: 1921–1925.[Abstract/Free Full Text]

7. Stauder R, Eisterer W, Thaler J, Gunthert U. CD44 variant isoforms in non-Hodgkin’s lymphoma: a new independent prognostic factor. Blood 1995; 85: 2885–2899.[Abstract/Free Full Text]

8. Ko LJ, Prives C. p53: puzzle and paradigm. Genes Dev 1996; 10: 1054–1072.[ISI][Medline]

9. Levine AJ. p53: the cellular gatekeeper for growth and division. Cell 1997; 88: 323–331.[ISI][Medline]

10. Giaccia AJ, Kastan MB. The complexity of p53 modulation: emerging patterns from divergent signals. Genes Dev 1998; 12: 2973–2983.[Free Full Text]

11. Evans SC, Lozano G. The Li-Fraumeni syndrome: an inherited susceptibility to cancer. Mol Med Today 1997; 3: 390–395.[ISI][Medline]

12. Soussi T. The p53 tumour suppressor gene: a model for molecular epidemiology of human cancer. Mol Med Today 1996; 2: 32–37.[ISI][Medline]

13. Lowe SW, Ruley HE, Jacks T, Housman DE. p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 1993; 74: 957–967.[ISI][Medline]

14. Lowe SW, Bodis S, McClatchey A et al. p53 status and the efficacy of cancer therapy in vivo. Science 1994; 266: 807–810.[ISI][Medline]

15. Brown JM, Wouters BG. Apoptosis, p53, and tumor cell sensitivity to anticancer agents. Cancer Res 1999; 59: 1391–1399.[Abstract/Free Full Text]

16. Falette N, Paperin M, Treilleux I et al. Prognostic value of p53 gene mutations in a large series of node-negative breast cancer patients. Cancer Res 1998; 58: 1451–1455.[Abstract]

17. Tada M, Matsumoto R, Iggo RD et al. Selective sensitivity to radiation of cerebral glioblastomas harboring p53 mutations. Cancer Res 1998; 58: 1793–1797.[Abstract]

18. Cote RJ, Esrig D, Groshen S et al. p53 and treatment of bladder cancer. Nature 1997; 385: 123–124.[ISI][Medline]

19. Sander CA, Yano T, Clark HM et al. p53 mutation is associated with progression in follicular lymphomas. Blood 1993; 82: 1994–2004.[Abstract]

20. Lo Coco F, Gaidano G, Louie DC et al. p53 mutations are associated with histologic transformation of follicular lymphoma. Blood 1993; 82: 2289–2295.[Abstract]

21. Gaidano G, Ballerini P, Gong JZ et al. p53 mutations in human lymphoid malignancies: association with Burkitt lymphoma and chronic lymphocytic leukemia. Proc Natl Acad Sci USA 1991; 88: 5413–5417.[Abstract]

22. Hernandez L, Fest T, Cazorla M et al. p53 gene mutations and protein overexpression are associated with aggressive variants of mantle cell lymphomas. Blood 1996; 87: 3351–3359.[Abstract/Free Full Text]

23. Dohner H, Fischer K, Bentz M et al. p53 gene deletion predicts for poor survival and non-response to therapy with purine analogs in chronic B-cell leukemias. Blood 1995; 85: 1580–1589.[Abstract/Free Full Text]

24. Drach J, Ackermann J, Fritz E et al. Presence of a p53 gene deletion in patients with multiple myeloma predicts for short survival after conventional-dose chemotherapy. Blood 1998; 92: 802–809.[Abstract/Free Full Text]

25. Ichikawa A, Kinoshita T, Watanabe T et al. Mutations of the p53 gene as a prognostic factor in aggressive B-cell lymphoma. N Engl J Med 1997; 337: 529–534.[Abstract/Free Full Text]

26. Koduru PRK, Raju K, Vadmal V et al. Correlation between mutation in p53, p53 expression, cytogenetics, histologic type, and survival in patients with B-cell non-Hodgkin’s lymphoma. Blood 1997; 90: 4078–4091.[Abstract/Free Full Text]

27. Wilson WH, Teruya-Feldstein J, Fest T et al. Relationship of p53, bcl2, and tumor proliferation to clinical drug resistance in non-Hodgkin’s lymphomas. Blood 1997; 89: 601–609.[Abstract/Free Full Text]

28. Chilosi M, Doglioni C, Magalini A et al. p21/WAF1 cyclin-kinase inhibitor expression in non-Hodgkin’s lymphomas: a potential marker of p53 tumor-suppressor gene function. Blood 1996; 88: 4012–4020.[Abstract/Free Full Text]

29. Maestro R, Gloghini A, Doglioni C et al. Human non-Hodgkin’s lymphomas overexpress a wild-type form of p53 which is a functional transcriptional activator of the cyclin-dependent kinase inhibitor p21. Blood 1997; 89: 2523–2528.[Abstract/Free Full Text]

30. Coiffier B, Gisselbrecht C, Herbrecht R et al. LNH-84 regimen: a multicenter study of intensive chemotherapy in 737 patients with aggressive malignant lymphoma. J Clin Oncol 1989; 7: 1018–1026.[Abstract]

31. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual, 2nd edition. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 1989.

32. Hamelin R, Jego N, Laurent-Puig P et al. Efficient screening of p53 mutations by denaturing gradient gel electrophoresis in colorectal tumors. Oncogene 1993; 8: 2213–2220.[ISI][Medline]

33. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

34. Cox DR. Regression models and life-tables. J R Stat Soc 1972; 34: 187–220.[ISI]

35. Mazars GR, Jeanteur P, Lynch HT et al. Nucleotide sequence polymorphism in a hotspot mutation region of the p53 gene. Oncogene 1992; 7: 781–782.[ISI][Medline]

36. Serra A, Gaidano GL, Revello D et al. A new TaqI polymorphism in the p53 gene. Nucleic Acids Res 1992; 20: 928.

37. Hainaut P, Hernandez T, Robinson A et al. IARC database of p53 gene mutations in human tumors and cell lines: updated compilation, revised formats and new visualisation tools. Nucleic Acids Res 1998; 26: 205–213.[Abstract/Free Full Text]

38. Avigad S, Barel D, Blau O et al. A novel germ line p53 mutation in intron 6 in diverse childhood malignancies. Oncogene 1997; 14: 1541–1545.[ISI][Medline]

39. Barel D, Avigad S, Mor C et al. A novel germ-line mutation in the noncoding region of the p53 gene in a Li-Fraumeni family. Cancer Genet Cytogenet 1998; 103: 1–6.[ISI][Medline]

40. Lehman TA, Haffty BG, Carbone CJ et al. Elevated frequency and functional activity of a specific germ-line p53 intron mutation in familial breast cancer. Cancer Res 2000; 60: 1062–1069.[Abstract/Free Full Text]

41. Cho Y, Gorina S, Jeffrey PD, Pavletich NP. Crystal structure of a p53 tumor suppressor–DNA complex: understanding tumorigenic mutations. Science 1994; 265: 346–355.[ISI][Medline]

42. Børresen-Dale AL, Lothe RA, Meling GI et al. TP53 and long-term prognosis in colorectal cancer: mutations in the L3 zinc-binding domain predict poor survival. Clin Cancer Res 1998; 4: 203–210.[Abstract]

43. Aas T, Børresen AL, Geisler S et al. Specific p53 mutations are associated with de novo resistance to doxorubicin in breast cancer patients. Nat Med 1996; 2: 811–814.[ISI][Medline]

44. Graeber TG, Osmanian C, Jacks T et al. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 1996; 379: 88–91.[ISI][Medline]

45. Kinzler KW, Vogelstein B. Life (and death) in a malignant tumour. Nature 1996; 379: 19–20.[ISI][Medline]

46. Fan S, El-Deiry WS, Bae I et al. p53 gene mutations are associated with decreased sensitivity of human lymphoma cells to DNA damaging agents. Cancer Res 1994; 54: 5824–5830.[Abstract]

47. Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature 1998; 396: 643–649.[ISI][Medline]

48. Smith ML, Fornace AJ. Genomic instability and the role of p53 mutations in cancer cells. Curr Opin Oncol 1995; 7: 69–75.[Medline]

49. Møller MB, Gerdes AM, Skjødt K et al. Disrupted p53 function as predictor of treatment failure and poor prognosis in B- and T-cell non-Hodgkin’s lymphoma. Clin Cancer Res 1999; 5: 1085–1091.[Abstract/Free Full Text]