Clonal Rearrangements and Malignant Clones in Peripheral T-cell Lymphoma

Full Text

INTRODUCTION Peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS) is a heterogeneous group of lymphomas with a mature immunophenotype of peripheral (postthymic) T-lymphocytes. This diagnosis covers more than 29% of T-cell lymphomas that do not belong to other nosological forms and is a diagnosis by exclusion [1, 2]. Clinically, the disease is aggressive (five-year overall survival rate is less than 32%), often advanced (69% of patients are diagnosed at stages III/IV) with extranodal sites being involved (bone marrow, skin, subcutaneous tissue, and lungs) [3]. It is believed that the morphological substrates of the tumors are T-lymphocytes of mature T-cells immunophenotype with a αβ-variant of the T-cell surface receptor (TCRαβ) and CD2+, CD3+, CD5+, CD7+, CD4+ or CD8+ markers, whose expression displays signs of aberrance (loss of one or more of them). Most often, PTCL-NOS has a CD4+/CD8-immunophenotype, less commonly a CD4-/CD8+ one. In some peripheral T-cell lymphomas, the expression of T-cell markers on the surface is limited, e.g. only CD2 or CD3. In addition, a small number of PTCL-NOS are γδ-T-lymphocytes lymphomas that cannot be classified as hepatolienal γδ-lymphomas or γδ-variant of large granular lymphocytes leukemia based on clinical and morphological data [3-6]. A study of clonal rearrangements of T-cell receptor genes in PTCL-NOS confirms T-cell clonality in complex diagnostic cases and proves the presence of a tumor [7-11]. The method consists of PCR amplification and analysis of the genes of the V-D-J-segments junction region of the T-cell receptors δ (TCRD), γ (TCRG), and β (TCRB). This region has a unique nucleotide sequence in each normal T-lymphocyte. A fragment analysis of amplification products derived from healthy tissue reveals a lot of peaks with a Gaussian distribution of their lengths (Fig.1A). Monoclonal samples with the same length of PCR products are present as a single peak (monoallelic rearrangement, Fig. 1B) or as two peaks (biallelic rearrangement, Fig. 1B). The length of the monoclonal PCR product is unique for the tumor clone and is identical in all affected tissues of a patient. Detection of a clone, for example, in bone marrow aspirate, indicates bone marrow involvement. Furthermore, the nature of rearrangements can reveal the degree of a lymphoid tumor maturity. Rearrangements during normal maturation of T-lymphocytes occur in sequence: TCRD (Vδ-Dδ, Dδ-Dδ, Dδ-Jδ, Vδ-Jδ) gene locus is rearranged first, followed by TCRG (Vγ-Jγ) and incomplete rearrangement of TCRB (Dβ-Jβ). Complete rearrangements of TCRB (Vβ-Jβ) and TCRα-locus (Vα-Jα) (Fig. 2) occur later [12, 13]. Since TCRδ (TCRD) genes are located inside TCRα-locus, they are cut out during TCRα rearrangement. Therefore, the entire range of clonal rearrangements observed in lymphomas can be divided into early and more mature ones. If a tumor contains TCRD and TCRG loci clonal products, incomplete rearrangement of Dβ-Jβ, and no complete Vβ-Jβ rearrangements of TCRB genes, this indicates an early nature of the rearrangements, which corresponds to a tumor of γδ-T-lymphocytes. More commonly, a tumor has a more mature spectrum of rearrangements: Vγ-Jγ, Dβ-Jβ, and Vβ-Jβ rearrangements present simultaneously, which is typical of most TCRαβ-lymphomas, including PTCL-NOS. According to published data, clonal rearrangements of the TCRG and TCRB genes are present in 81-94% and 96% of PTCL-NOS, respectively [7, 8]. MATERIALS AND METHODS A retrospective analysis of clonality studies results in PTCL-NOS over the last 10 years (2005-2015) has been conducted in the National Hematology Research Center of the Russian Ministry of Health (hereinafter the NHRC). Patients and samples The set of patients consisted of 30 people (15 men and 15 women, median age 56 years (32-75)). The disease stage was determined according to the Ann-Arbor classification (1971); bone marrow involvement was considered to be stage IV. Four patients were diagnosed with stage I; four, with stage III; and 22, with stage IV. Lymph nodes were involved in 18 (60%) patients; bone marrow, in 22 (73%); spleen, in 14 (47%); skin, in 5 (17%); gastrointestinal tract, in 4 (13%); lungs, in 4 (13%); tonsils, in 3 (10%); liver, in 3 (10%); mediastinum, in 2 (7%); meninges (neuroleukemia), in 2 (7%); mammary gland, in 2 (7% ); and soft tissues of the orbit, in 1 (3%) (Table 1). Histological and immunohistochemical studies were performed in the NHRC Department of Pathology; and molecular and genetic studies of clonality, in the NHRC Laboratory of Molecular Hematology. Isolation of DNA from tissues Leukocytes and DNA from blood and bone marrow samples were isolated as described [14]. For isolation of DNA from tissue embedded into a paraffin block, five 5-μm sections were collected in Eppendorf tubes. The tissue was dewaxed by heating [15, 16]. Freshly frozen tissue for DNA extraction was thawed, and a 1×1×1 mm piece was cut out. DNA was isolated by the method based on tissue dissolution in concentrated ammonia with subsequent neutralization with glacial acetic acid and salting-out of proteins [17]. DNA concentration was determined using a UV spectrophotometer. DNA samples were stored at -20°C. Studies of TCR gene rearrangements by PCR and fragment analysis T-cell clonality was assessed using multiplex BIOMED-2 primers systems for fragment analysis [13] based on the rearrangements of the TCRG (Vγ-Jγ) and TCRB (Vβ-Jβ, Dβ-Jβ) genes. In the case of γδ-Tcell lymphomas, TCRD genes rearrangements were also analyzed. Multiplex amplification of TCRD genes was performed in duplicate tubes according to the BIOMED-2 protocol, Tube A and Tube B, and TCRB gene amplification, in three tubes, Tube A, Tube B, and Tube C (see description of the reactions in Table 2). Primers produced by Syntol (Russia) were used for amplification of the TCRD, TCRG genes. The reaction mixture in a final volume of 20 μl included 100 ng of DNA, 10 μl of a 2 × PCR mixture (PCR Master Mix Promega), and 5 pmol of each primer. TCRB genes were amplified using a commercial TCRB Gene Clonality Assay ABI Fluorescence Detection kit (Invivoscribe Technologies) and AmpliTaq Gold DNA polymerase (Applied Biosystems) according to the manufacturers’ instructions. PCR conditions were as follows: pre-denaturation at 95°C (5 min); 35 cycles at 92°C (35 seconds), 60°C (35 seconds), 72°C (35); and final elongation at 72°C (10 min). PCR was performed using an automated DNA Engine thermocycler (BioRad, Hercules, USA). Statistical analysis To compare the results obtained by the two methods, the Spearman rank correlation coefficient was calculated using the following formula: rs = 1 - 6Σd2/(N3 - N), where N is the sample size; d is the difference between the ranks for each member of the sample; and rs is the Spearman coefficient. RESULTS TCR gene clonality was detected in 29 of the 30 patients (97%): for TCRG genes, in 27 of the 30 (90%); and for TCRB genes, in 29 of the 30 (97%). In some PTCL samples, the rearrangement was detected in only one of the loci, TCRG or TCRB, which is due to abnormal differentiation of tumor cells and is often associated with an immature and aberrant immunophenotype. For example, in patient 27, whose bone marrow had only one clonal Dβ-Jβ rearrangement, the surface of the tumor cells did not have CD3, CD5, CD4, or CD8, but only CD2 and CD7. The most likely explanation for the lack of clonal peaks in the study of TCR genes (patient 25) is the small number of tumor cells in the sample (many reactive T-lymphocytes). PCR detected bone marrow involvement or the presence of clonal lymphocytes in the blood in 93% (25 out of 27) of patients (Table 1). The clonal rearrangements in bone marrow were not detected only in patients 25 and 26, for whom clonal peaks were also absent or doubtful in the lymph nodes. Morphological methods failed to detect bone marrow involvement in four patients (№ 16-19), whereas PCR detected clonal cells in their bone marrow. Detection of clonal rearrangements of TCRG genes in the bone marrow or blood is considered to be a poor prognostic factor in PTCL [18]. In most patients, we observed multiple (more than two) clonal rearrangements in one locus. Tumor lymphocytes clone may have a rearrangement on one chromosome (monoallelic rearrangement) or on two homologous chromosomes (biallelic rearrangement). Therefore, only one or two clonal peaks can be detected for each gene (TCRD or TCRG or TCRB) in one tumor clone. Additional Dβ2-Jβ rearrangement (Fig. 3) is described as an exception [19, 20]. In this case, an additional clonal peak in a range from 170-210 bp was observed in Tube C during fragment analysis after amplification of TCRB genes (Table 2). In 13 out of the 30 (43%) patients, three or more clonal peaks were detected in one TCR gene locus in at least one tissue (10 patients, in the TCRG locus; 11, in the TCRB locus). Multiple (three or more) clonal peaks were detected at the same frequency in bone marrow, lymph nodes, and/or spleen (in 10 out of 13 patients). In theory, the “extra” peaks can be attributed to reactive T-cells, but we faced a completely different situation. In 63% of the patients (15 out of 24) in whom we had analyzed several tissues, we observed the appearance of new clonal peaks and new clones in various tissues. The clonal rearrangements reported earlier were either not detected or partially preserved. There was a correlation between observations for the TCRG and TCRB genes: i.e., the appearance of a new clonal peak of TCRG genes was usually accompanied by the identification of a new clonal rearrangement of TCRB genes. This pattern can only be explained by the presence of several tumor clones and their different representations in the tissues and organs. In total, several clones were detected in 19 of the 30 patients (63%). The number of clones ranged from two to seven (Table 1). There was no correlation with age (p = 0.43) or the stage of the disease (p = 0.29). Identification of several clones correlated with the number of analyzed tissues (rs = 0.6, p < 0.0005). This phenomenon might have been overlooked in other patients due to the low number of investigated tissues. The most representative examples of clonality studies in PTCL are given below. Patient 11 (Fig. 4). At the time of the diagnosis, Patient 11 displayed a pattern typical of mature T-cell lymphoma: biallelic rearrangement of TCRG genes (212 and 224 bp) and complete rearrangement of TCRB genes (Tube B, 262 bp). Bone marrow examination revealed two more clones with incomplete rearrangement of TCRB genes (Tube C 187 and 193 bp). Progression of the disease leads to skin lesions, which were accompanied by one or two new clonal rearrangements of TCRG genes (201 bp) and complete rearrangement of TCRB genes (Tube A 254 bp). Further progress leads to the development of neuroleukemia. The cerebrospinal fluid (CSF) lacked the clones previously detected in the tumor (TCRG 212 and 224 bp) but contained a clone with rearrangement of TCRG genes, 201 bp in length, and a new clone with rearrangement of TCRG genes, 174 bp in length. After 11 months, the patient underwent splenectomy and liver biopsy due to growing cytopenia. The clonal rearrangement pattern in spleen and liver is identical to the original one in the stomach and lymph node. Therefore, Patient 11 has a total of four rearrangements of the TCRG gene: two full rearrangements of TCRB genes and two incomplete rearrangements of the TCRB gene. The dynamics of appearance of new rearrangements during the progression of the disease indicates five or more different tumor clones. Patient 9 (Fig. 5). Patient 9 had multiple clonal rearrangements of the TCRG and TCRB genes with a varying presence of clones in different tissues and organs. The primary diagnosis included examination of the blood, bone marrow, tonsil biopsy and lymph node 1. Five or more rearrangements of the TCRG and TCRB genes were detected in the blood and bone marrow. Tonsil and lymph node 1 contained three and four clonal rearrangements of TCRG, respectively, and four rearrangements of the TCRB gene. Lymph node 2 and skin biopsies were performed with progression of the disease. The lymph node biopsy revealed five rearrangements of the TCRG gene and eight of the TCRB genes. Except for the rearrangements of TCRB (A), 267 bp, and TCRB (B), 262 bp, that are present in most tissues simultaneously, other rearrangements of TCRB genes were randomly distributed and belonged to different tumor clones. The data of molecular clonality studies indicate the presence of seven or more tumor clones in this patient. Patient 8 (Fig. 6). Patient 8 had five different rearrangements of TCRG genes (177, 183, 204, 225, 248 bp) with different clones present in different studied tissues. We found four rearrangements of TCRB: three complete rearrangements (Tubes A and B) and one incomplete Dβ2-Jβ2 (Tube C). In this case, there was no clear connection between the TCRG and TCRB genes rearrangements. For example, the lymph node 1 and skin predominantly had TCRG gene rearrangements, 183 and 248 bp in size, however, the pattern for TCRB was different. The number of clones in this patient was four or higher, but it is possible that the number of clones could be much higher. For example, the clones with only a TCRG or TCRB rearrangement, or clones with rearrangements of TCRG 177 bp and TCRB 260 bp, TCRG 177 bp and TCRB 189 bp, etc. DISCUSSION The presence of several tumor clones described in acute lymphoblastic leukemia (ALL) is attributed to an “ongoing” process of rearrangement of immunoglobulins and TCR genes in early progenitor cells [21-24]. Clonal products sequencing during ALL manifestation and in relapse showed that clones with incomplete rearrangements of the TCRD and TCRB genes and their derivatives with complete clonal rearrangements are simultaneously present in ALL. Furthermore, in some cases complete rearrangement is modified. The V gene is replaced with another (upstream) gene or the rearrangement is completely replaced by another one of the upstream V and downstream J genes, i.e. by the genes that are distal to the previous rearrangement. In some cases, there is a deletion and disappearance of the TCR gene rearrangement in relapse. We believe that the molecular events that occur in ALL can also occur in PTCL. In tumor cells, the control mechanisms which are responsible for preventing further restructuring of a locus after a productive rearrangement are disrupted, which leads to increased activity of the RAG1 and RAG2 enzymes, changes in chromatin organization, etc. It is possible that in PTCL rearrangements can be replaced with new ones and/or that an incomplete Dβ-Jβ one can be replaced with a complete Vβ-Jβ one. In addition, there is general chromosomal instability, which can lead to deletions or duplications of the TCR genes locus. The deletion of the locus may trigger further rearrangements on homologous chromosomes. Complex chromosomal changes, including triploidy, tetraploidy, loss of chromosomes, 7q trisomy, and translocations involving TCR genes loci (14q11, 7q34-35, 7p13-21) are described in various peripheral T-cell lymphomas [25-27]. In any case, clonal rearrangements are common markers that indicate the heterogeneity of the tumor and clonal evolution during the progression of the disease. We have observed “multiclonality” in most patients, but it is unclear whether this phenomenon is present only in PTCL. It is possible that some other lymphomas also have high clonal heterogeneity, but their “visualization” requires different approaches and methods. CONCLUSIONS Studies of TCRG and TCRB genes clonality effectively prove the presence of a tumor in the majority (97%) of PTCL patients. PCR revealed bone marrow involvement and/or presence of clonal lymphocytes in the blood in the majority (93%) of patients, whereas in morphological studies bone marrow involvement was confirmed only in 73% of the patients. A specific pattern of rearrangements in PTCL (multiple rearrangements of the TCRG and TCRB genes, loss and gain of new ones, presence of several clones of a tumor) was observed in the majority of patients (63%) and should certainly be taken into account when using this method for diagnostic purposes. The emergence of new clonal peaks (clones) should not be considered to be the emergence of a new tumor. In addition, multiple rearrangements may interfere with minimal residual disease monitoring in PTCL.

About the authors

Yu. V. Sidorova

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

N. G. Chernova

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

N. V. Ryzhikova

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

S. Yu. Smirnova

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

M. N. Sinicina

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

Yu. E. Vinogradova

I.M. Sechenov Moscow State Medical University

Email: dusha@blood.ru
Russian Federation

H. L. Julhakyan

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

A. M Kovrigina

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

E. E. Zvonkov

National Hematology Research Center of the Ministry of Health

Email: dusha@blood.ru
Russian Federation

A. Б. Sudarikov

National Hematology Research Center of the Ministry of Health

Author for correspondence.
Email: dusha@blood.ru
Russian Federation

References

Supplementary files