This article is mentioned in:

Introduction

Myelodysplastic syndromes (MDS) are a group of clonal hematological disorders characterized by ineffective hematopoiesis, peripheral cytopenias and an increased risk of development of acute myeloid leukemia (1). A number of chromosomal abnormalities have been detected in ~50–60% of patients with de novo MDS and in ≤80% of patients with therapy-related MDS (2). Among the classical oncogenic abnormalities, those involving the tumor protein 53 (TP53) gene have been extensively investigated in MDS. TP53 mutations were detected primarily in high-risk and therapy-related MDS, frequently in association with complex chromosomal abnormalities (3,4). In such diseases, TP53 mutations were revealed to have a negative prognostic impact (3,5).

The wild-type p53 protein is normally undetectable by immunohistochemistry (IHC), due to its short half-life. By contrast, the majority of mutated proteins have a prolonged half-life, accumulate within the nucleus and can be easily detected in formalin-fixed, paraffin-embedded tissues (6). Thus, the immunohistochemical detection of p53 protein suggests an underlying mutation in the gene (7). It has been previously demonstrated that aberrant nuclear expression of p53 protein is associated with hemizygous p53 deletion in multiple myeloma (8,9) and chronic lymphocytic leukemia (10). In MDS, immunohistochemical staining for TP53 protein in bone marrow trephine biopsies has revealed a strong correlation with TP53 mutation status (11–13).

A large number of previous studies have reported on the strong prognostic significance of the immunohistochemical detection of p53 protein in tumor pathology; however, few studies have evaluated its prognostic value in MDS (14–16), and certain studies primarily focused on low-risk MDS with del(5q) (11,17).

In the present study, p53 immunoreactivity (p53-IR) was investigated in bone marrow biopsies (BMBs) of patients with MDS that underwent bone marrow transplantation (BMT), with the aim of determining its association with clinical and histopathological data.

Patients and methods

A total of 18 patients that were admitted to the Division of Haematology (Città della Salute e della Scienza and University of Turin, Italy) from December 1994 to June 2005, with a diagnosis of MDS were retrospectively examined. There were 6 females and 12 males; the mean age was 50.5 years (range, 32–64). Diagnosis of MDS was performed according to the World Health Organization (WHO) criteria (1). In total, 5 patients had refractory cytopenia with multilineage dysplasia (RCMD) and 13 patients had refractory anemia with excess blasts, type 2 (RAEB-2). Hemoglobin level, white blood cell (WBC) and platelet counts were assessed from peripheral blood at diagnosis. All patients underwent allogeneic bone marrow transplantation. General informed consent was obtained according to the guidelines of the Ethics Committee, Città della Salute e della Scienza and University of Turin. Samples were numerically identified, maintaining patients' anonymity.

BM morphology and immunohistochemistry

Serial sections (3-µm thick) from Bouin's solution-fixed, paraffin-embedded BMBs were stained with hematoxylin-eosin, Dominici, Perls and reticulin (Fig. 1A) and immunostained with an automated stainer device (Leica BOND III, Leica Biosystems, Melbourne Pty Ltd, Mount Waverley VIC 3149, Australia) using a polyclonal antibody against myeloperoxidase (cat. no. A0398; Dako; Agilent Technologies, Inc., Santa Clara, CA, USA; dilution, 1:1,000), and monoclonal antibodies against glycophorin A (clone JC159, #M0819; Dako; Agilent Technologies, Inc., Santa Clara, CA, USA; dilution, 1:50), CD61 (clone 2f2; cat. no. 760-4249; Ventana Medical Systems, Tucson, AZ, USA; undiluted), CD34 (Fig. 1B) (clone QBEnd/10; cat. no. NCL-L-END; Novocastra; Leica Microsystems, Milton Keynes, UK; dilution, 1;50) and p53 (clone DO7, #NCL-L-p53-DO7; Novocastra; Leica Microsystems, Milton Keynes, UK; dilution, 1:100) at room temperature for 15 min.

Figure 1. (A) RAEB-2 exhibiting moderate and diffuse reticulin fibrosis (MF-2) (bone marrow biopsy; Gomori stain; magnification, ×400). (B) RAEB-2 showing numerous CD34 immunopositive blasts arranged in clusters (bone marrow biopsy; CD34 immunoperoxidase staining; magnification, ×400). (C) RAEB-2 in which >50% of cells have intensely stained nuclei for p53 (p53-IR positive case; bone marrow biopsy; p53 immunoperoxidase staining; magnification, ×400). (D) Refractory cytopenia with multilineage dysplasia. Only one myeloid cell shows an intensely p53 stained nucleus (p53-IR negative case) (bone marrow biopsy; p53 immunoperoxidase staining; magnification, ×400). RAEB-2, refractory anemia with excess blasts, type 2; p53-IR, p53 immunoreactivity.

Paraffin sections (3-µm thick) on SuperFrost microscope slides were processed with an automated stainer device (Leica BOND III, Leica Microsystems) and pretreated for 30 min in citrate buffer (pH 6.0; Bond Epitope Retrieval Solution 1, Leica Microsystems). The DO-7 antibody was applied at a 1:100 dilution for 30 min at room temperature and detected using a Bond Polymer Refine Detection kit (cat. no. DS9800; Leica Microsystems) according to the manufacturer's protocol. The percentage of cells with intense (p53++) nuclear staining was calculated examining ≥1,000 hematopoietic cells at a high magnification (x40) using a standard light microscope. The cut off for positivity (13) was 5% of stained cells (Fig. 1C and D).

Statistical analysis

The association between p53 immunoreactivity (p53-IR) and clinical or hematological parameters was assessed using one-way analysis of variance and the Fisher's exact test. Univariate survival analyses were based on Kaplan-Meier product-limit estimates of survival distribution, and differences between survival curves were tested using the log-rank test. Overall survival (OS) was calculated from the date of diagnosis to the date of the last observation or death. All analyses were performed using SPSS version 17 (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Results

Association between p53-IR, clinical and hematological features and bone marrow histology

Positive p53-IR was detected in 7/18 (38.9%) patients. Significant associations were detected between p53-IR and patient age (mean age, 57.3 years for positive p53-IR patients vs. 46.4 years for negative p53-IR patients; P=0.005) and the pattern of BM fibrosis, in which, of the 8 patients with diffuse fibrosis, 6 (75%) were p53-IR-positive compared with 1 (16.7%) of the 6 patients with focal fibrosis (P=0.03). A trend toward significance was found for sex, (positive p53-IR in 67% of females vs. 25% of males; P=0.08), BM cellularity (mean BM cellularity, 81.4% for p53-IR-positive patients vs. 64.5% for p53-IR-negative patients; P=0.1) and the degree of BM fibrosis (44.4% of WHO-MF1 and 60% of WHO-MF2 cases were p53-IR-positive, whereas no WHO-MF0 cases were positive; P=0.1). No association was found for hemoglobin level, white blood cell (WBC) and platelet counts, percentage of CD34 positive blasts and MDS type; however, the rate of positive p53-IR was higher in patients with RAEB2 (46.2%) than in those with RCMD (20%). The results are summarized in Table I.

Table I. Association between p53-IR and clinical and histopathological bone marrow features.

Table I.

Association between p53-IR and clinical and histopathological bone marrow features.

		P53-IR positive (N=7)	P53-IR negative (N=11)
Variable	N	Mean ± SD	Mean ± SD	P-valuea
Age, years	18	57.3±4.6	46.4±8	<0.01
Hb level, g/dl	18	8.38±1.8	9.06±1.70	0.40
WBC count, ×109/l	18	3.421±2.989	4.380±3.130	0.60
Plt count, ×109/l	18	88.7±99.1	119.5±81.1	0.50
BM cellularity, %	18	81.4±10.7	64.5±27.3	0.10
BM blasts, %	18	13.1±5.9	11.4±7.7	0.60
Variable	Ν	N (%)	N (%)	P-valueb
Sex
Male	12	3 (25)	9 (75)
Female	6	4 (66.7)	2 (33.3)	0.08
MDS type
RCMD	5	1 (20)	4 (80)
RAEB-2	13	6 (46.2)	7 (53.8)	0.30
BM fibrosis WHO grade
MF-0	4	0 (0)	4 (100)
MF-1	9	4 (44.4)	5 (55.6)	0.10
MF-2	5	3 (60)	2 (40)
BM fibrosis pattern
Focal	6	1 (16.7)	5 (83.3)
Diffuse	8	6 (75)	2 (25)	0.03

a One-way analysis of variance

b Fisher's exact test. p53-IR, p53 immunoreactivity; WBC, white blood cell; Plt, platelet; BM, bone marrow; RCDM, refractory cytopenia with multilineage dysplasia; RAEB, refractory anemia with excess blasts; SD, standard deviation; Hb, hemoglobin; MDS, myelodysplastic syndromes; WHO, World Health Organization.

Associations between p53-IR, clinicopathological parameters and overall survival

At the time of analysis, 13 mortalities (72% of patients) had occurred, and 5 patients (28%) were alive (censored). The mean OS for the whole series was 73 months (median, 18; range, 2–216). OS was significantly associated with patient age: At 5-year follow-up, 55% of patients younger <51 years were alive compared to 44% of older patients (P=0.01). OS was also associated with hemoglobin level (5 year OS was 73% for patients with hemoglobin >8 g/dl vs. 14% for patients with hemoglobin ≤8 g/dl; P=0.04), the type of MDS (5 year OS was 80% for patients with RCMD vs. 38% for patients with RAEB-2; P=0.05), the degree of BM fibrosis (Fig. 1A) (all patients without fibrosis were alive after 5 years vs. 34% of patients with WHO MF1/2; P=0.006), the number of BM CD34-positive blasts (Fig. 1B) (80% of patients with <5% BM blasts were alive after 5 years vs. 38% of those with >5% BM blasts; P=0.05). The median survival for patients with a negative p53-IR (Fig. 1D) was 105 months vs. 8 months for patients with a positive p53-IR (Fig. 1C) (P=0.1). No difference in OS was found for sex, WBC and platelet counts. The results are summarized in Table II.

Table II. Correlation between clinical and histopathological bone marrow features and p53-IR with overall survival.

Table II.

Correlation between clinical and histopathological bone marrow features and p53-IR with overall survival.

Variable	N	Median overall survival, months	1-year overall survival rate, %	5-year overall survival rate, %	P-valuea
Whole series	18	18	61	49
Sex
Male	12	75	67	58
Female	6	8.2	50	33	0.60
Age, years
≤51	9	108	66	55
>51	9	13	55	44	0.01
WBCs, ×109/l
≤1.8	5	12	60	40
>1.8	11	9	55	44	0.70
Hb level, g/dl
≤8	7	8	42	14
>8	11	105	73	73	0.04
Plt count, ×109/l
≤100	10	75	70	60
>100	8	9	50	37	0.80
MDS type
RCMD	5	105	100	80
RAEB-2	13	12	46	38	0.05
BM fibrosis WHO grade
MF-0	4	n.a.	100	100
MF-1/2	14	12	57	34	<0.01
BM fibrosis pattern
Focal	6	18	67	50
Diffuse	8	8.2	37.5	25	0.20
BM CD34 positive blasts
≤5	5	105	100	80
>5	13	12	46	38	0.05
P53-IR
Negative	11	105	73	55
Positive	7	8	57	43	0.10

a Log-rank test. WBC, white blood cell; Plt, platelet; BM, bone marrow; RCDM, refractory cytopenia with multilineage dysplasia; RAEB, refractory anemia with excess blasts; p53-IR, p53 immunoreactivity; Hb, hemoglobin; MDS, myelodysplastic syndromes; WHO, World Health Organization; n.a., not applicable.

Discussion

Intense p53 (++) immunoreactivity was detected in 7/18 cases (38.8%), in line with the rate of positivity (44%) previously reported in RCMD/RAEB2 cases (13) using similar staining and scoring procedure. Although previous studies used 1 or 2% of intensely immunostained nuclei as a cut off for p53 positivity (11,17), the present study used 5% strongly (++) immunopositive nuclei as a cut off, as it has been demonstrated that in 55 del(5q) patients, all those with >5% p53++ stained cells in BM biopsies carried the TP53 mutation (11).

p53 immunopositivity was associated with BM fibrosis in the current study, in which no patients without BM fibrosis were p53-IR positive, whereas 50% of those with WHO MF1-2 were, and 75% of patients with diffuse reticulin fibrosis were p53-IR positive, by contrast to only 16% of patients with focal fibrosis (P=0.03). Although no significant association was detected, p53 positivity was higher in RAEB2 than RCMD, in accordance with large studies revealing that TP53 mutations were observed mainly in patients with intermediate-2 or high risk MDS (5). Therefore, with the limitation due to the number of cases, the current results indicate that a positive p53-IR at diagnosis is associated with adverse histological prognostic factors, such as BM fibrosis, and with clinically more aggressive MDS subtypes.

In addition, OS was associated with fibrosis in the present study; all patients without evidence of fibrosis in BM biopsy were alive at the 5-year follow-up, whereas only 34% of those with WHO MF-1/2 disease were (P=0.006). This finding is in agreement with large studies demonstrating the clinical relevance of bone marrow fibrosis in primary myelodysplastic syndromes (18,19). OS was also associated with the type of MDS: 80% of patients with RCMD were alive at the 5-year follow-up, whereas only 38% of those with RAEB2 were (P=0.05).

Notably, in the present series, a positive p53-IR tended to be associated with a shorter OS: The median survival for patients with negative p53-IR was 105 months but only 8 months for those with positive p53-IR (P=0.1). This result is concordant with studies revealing the strong prognostic value of p53 mutations in MDS (3,5,12,20). Mutations in TP53, enhancer of zeste homolog 2, ETS variant 6, runt-related transcription factor 1 and additional sex combs like 1 transcriptional regulator proteins were found to be predictors of poor overall survival in patients with MDS; however, only mutations in the TP53 gene have been clearly associated with poor prognostic markers, and have been reported to independently predict survival, primarily in patients with intermediate-2 or high risk MDS (5). A large single institution study on 318 patients with MDS revealed that TP53 mutations were the strongest predictor of outcome in a multivariate model (12).

The present results emphasize the prognostic value of the immunohistochemical detection of p53 protein in MDS (11,14–17). The finding is particularly relevant, when considering the small number of cases and the type of therapy of the present series. All patients underwent allogeneic BM transplantation, by contrast to the previously reported series, which were primarily concerning patients with low- or intermediate risk del(5q) MDS treated with lenalidomide.

Immunohistochemistry is a practical and convenient method to detect p53 protein overexpression in tumor cells (6). An association has been reported between intense p53 nuclear staining and TP53 mutation in MDS (11,13,17); furthermore, patients without TP53 mutations did not exhibit intense p53 protein staining, conferring a good negative predictive value for IHC (12). Therefore, the acceptable sensitivity of p53 immunostaining in predicting for TP53 mutations is encouraging, and the methodology is routine in diagnostic laboratories.

In conclusion, our findings suggest that the evaluation of p53-IR on BMBs of patients with MDS may be introduced in the histopathological work-up of the disease.

Acknowledgements

The present study was supported by grants from the Ministero Italiano dell'Università e Ricerca Scientifica (MIUR ex 60%). This study was presented in part at the 19th World Congress on Advances in Oncology and 17th International Symposium on Molecular Medicine, 9–11 October 2014, Athens, Greece.

References