Tyrosine kinase inhibitors are a family of oral, small-molecule drugs that are targeted against the oncogenic fusion protein BCR-ABL, which is formed when the ABL gene on chromosome 9 joins to the BCR gene on chromosome 22, leading to the characteristic t(9;22) known as the Philadelphia chromosome(1-3). This is the pathognomonic translocation in CML, but can also be found in 20% of adult ALL and 50% of elderly ALL patients(4).
The first generation BCR-ABL TKI is Imatinib (Gleevec), with second-generation BCR-ABL Dasatinib (Sprycel), Nilotinib (Tasigna) and Bosutinib (Bosulif) which are several orders of magnitude more potent than Imatinib. Nilotinib, for example, is a close analog of Imatinib, but it has a 20-fold higher potency for BCR-ABL kinase inhibition than Imatinib, whereas Dasanitib has yet another 10-fold increased potency compared to Nilotinib(5,6). In addition, novel kinase and non-kinase targets have been described in these TKI’s, and Dasatinib in particular was developed as an immunosuppressor, with dual specificity against BCR-ABL and SRC-kinases(7,8).
Dasatinib’s unique SRC-kinase and TEC family kinase activities are thought to contribute to its immunomodulatory properties, as these proteins play a role in the signaling pathways of the T- and B-cell receptors(7). One such immunomodulatory phenomenon that has been observed is the appearance of Large Granular Lymphocytosis (LGL) in the peripheral blood of approximately 30% of patients treated with this drug(9-14). Furthermore, the patients with LGL were also noted to an increased incidence of autoimmune-mediated side effects such as pleural effusions, colitis, and fevers(8,11,15).
In this paper, we describe the case of a Philadelphia-positive Acute Lymphoblastic Leukemia patient who was treated with Dasatinib, developed LGL and pleural effusions, and had a favorable clinical response.
Mr. V.S. is a 73-year-old male who initially presented to his primary care physician with dyspnea on exertion and lower extremity edema.
A complete blood count done by his primary care physician revealed a white blood cell count (WBC) of 50,000/ul (10% neutrophils, 1% myelocytes, and 89% blasts), a hemoglobin concentration of 5.9 g/dl, and a platelet count of 181,000/ul, and based on it he was referred for hospital admission.
The blood chemistry data obtained on admission were as follows: sodium of 139 mmol/L, potassium of 4.3 mmol/L, chloride of 103 mmol/L, carbon dioxide of 24 mmol/L, blood urea nitrogen of 24 mg/dl and a creatinine of 1.27 mg/dl. The liver function testing showed a total protein of 7.1g/dl, serum albumin of 3.7g/dl, aspartate aminotransferase (AST/SGOT) of 40 U/L, alanine aminotransferase (ALT/SGPT) of 21 U/L, total bilirubin of 0.3mg/dl, and an alkaline phosphatase of 86 U/L. The lactate dehydrogenase level was significantly elevated at 1056 U/L (normal range is between 50 and 242 U/L), with a uric acid of 8.4 mg/dl. The coagulation profile showed an activated partial thromboplastin time of 24.7 seconds, a prothombin time of 13.9 seconds with a corresponding INR ratio of 1.23, a fibrinogen of 478 mg/dl and an elevated D-dimer assay of 1067 ng/ml (normal range is between 0 and 499 ng/ml).
The patient received packed red blood cell transfusions for the anemia which led to symptomatic improvement.
The physical exam was significant for lower extremity edema bilaterally, with no lymphadenopathy or hepatosplenomegaly. Computer tomography (CT) scanning of the chest/abdomen/pelvis showed a left lower lobe pneumonia which was treated with intravenous antibiotics, but there was no evidence of lymph node enlargement, mediastinal masses, or hepatic/splenic enlargement.
A bone marrow biopsy was performed, which showed a hypercellular marrow (up to 100% cellularity) with an extensive infiltrate of immature cells, scattered erythroid precursors and megakaryocytes with normal morphology.
The flow cytometry studies showed a lymphoblast population (78% of cells), positive for HLA-DR, CD38, (partial) CD34, CD19, CD10, CD22, (partial) CD20, (partial) CD15 and negative for CD13, CD 33, CD117 and surface kappa and lambda light chain determinants.
The blasts were positive for TdT by immunofluorescence staining.
The findings were consistent with B-lymphoblastic leukemia.
Chromosomal analysis showed an abnormal karyotype: eleven out of twenty examined metaphases had the t(9;22) Philadelphia (Ph+) chromosome, nine metaphases had a normal 46XY male karyotype, and one of the eleven Ph+ metaphases had additional copies of chromosomes 6, 8, 12, 14, 21 making up for a total of 54 XY.
Florescence in situ hybridization (FISH) was also positive for BCR/ABL1 (9q34/22q11.2) rearrangement in 83.5% of cells.
The patient was consented and enrolled in a phase II multi-institutional clinical trial of Dasatinib as a primary treatment in Ph+ ALL adults over the age of 50 (Cancer and Leukemia Group B, CALGB 10701, NCT01256398).
A pre-treatment Multiple-Gated Acquisition (MUGA) scan showed a normal ejection fraction of 60%. His induction course consisted of continuous Dasatinib 140 mg orally daily in combination with Dexamethasone 10 mg/m2/day orally for days 1-7, and given a good response on the day 15 bone marrow biopsy (less than 20% marrow blasts), the patient continued treatment on single agent Dasatinib treatment.
After 20 days, the patient developed shortness of breath, and a chest X-ray showed new bilateral pleural effusions. There was no change to his ejection fraction on a trans-thoracic echocardiogram, and the effusions were attributed to Dasatinib. He was treated with the diuretic Furosemide, and due to persistence of symptoms, Dasatinib was briefly interrupted and the patient underwent a thoracoscopy which effusion drainage. The dose of his Dasatinib was resumed at 100 mg orally daily, which he then tolerated well. A day 30 bone marrow biopsy showed complete morphological remission as well as a normal karyotype, 46XY and FISH negative for BCR-ABL1.
As an outpatient, he continued on oral Dasatinib 100 mg daily with Furosemide for residual lower extremity edema and small pleural effusions, and his complete blood counts were drawn at weekly intervals. He was noted (after about two months from the initiation of Dasatinib) to have an elevated WBC count of 10.4K/ul, with an absolute neutrophil number of 3400/ul, absolute lymphocyte number of 6200/ul (normal range is between 1000/ul and 3300/ul), absolute monocyte count of 800/ul, 100/ul eosinophils and 0 basophils. Peripheral smear showed an increased number of large granular lymphocytes, with no blasts present.
The hemoglobin was 10.8 g/dl, with a platelet count of 306k/ul.
Peripheral blood was sent for flow cytometry and it identified, by morphology, 42% lymphocytes and 36% large granular lymphocytes.
The immunophenotype showed Natural-Killer (NK)-like T cells (19% of cells) positive for CD2, CD3, CD5, CD7, CD8, CD56, CD57 and TCR alpha/beta.
Natural killer cells were also present (10% of cells) and T-cells, with decreased CD4 to CD8 ratio (CD4% was 16, CD8% was 72, with CD4/8 ratio of 0.23 - normal range is between 0.90 and 3.60). Molecular studies for T-cell receptor (TCR) gene rearrangement identified discrete bands in both TCR-beta and TCR-gamma analysis, suggestive of a clonal T-cell population.
The patient had no lymphadenopathy, hepatosplenomegaly or dyspnea.
As a result of a low-grade fever (100.8 degrees Fahrenheit), blood cultures, urine cultures, chest X-rays and Cytomegalovirus (CMV) titer via Polymerase Chain Reaction (PCR) were checked. The CMV PCR titer was low-positive at 500; this was monitored with no antiviral therapy, and it became negative approximately two weeks later.
The patient was monitored clinically, and his lymphocytosis was noted to gradually decrease over the course of three weeks to normal, and not to recur despite the continued administration of Dasatinib at the same dose. He achieved a deep disease response at six months of treatment, namely bone marrow molecular complete response evidenced by a negative BCR-ABL by Polymerase Chain Reaction (PCR). He is currently on continuous maintenance phase, which includes Dasatinib in addition to oral Methotrexate, oral 6-Mercaptopurine and monthly intravenous Vincristine and oral Dexamethasone, and has maintained a bone marrow morphological remission at a year and a half since diagnosis.
Large Granular Lymphocytes (LGL) generally represent 10-15% of the circulating mononuclear cells in adults, with most having an NK-cell phenotype (CD3-, CD16+, CD56+)(9). A clonal increase in the number of peripheral blood LGL’s generally leads to the diagnosis of LGL Leukemia, an indolent lymphoproliferative disorder that is clinically characterized by lymphocytosis, cytopenias (most commonly neutropenia and anemia), hepatosplenomegaly, fever, and autoimmune conditions (most commonly rheumatoid arthritis)(16). LGL Leukemia, in the appropriate clinical setting (such as severe neutropenia or anemia), requires treatment with immunosuppressive agents, typically oral Methotrexate or oral Cyclophosphamide(17).
However, it has been recognized that both monoclonal and oligoclonal LGL expansions can occur after allogeneic stem cell transplantation, and in setting of chronic viral infections(18). When present, they are associated with long-term remission in leukemic patients(19).
Similarly, LGL has been described in around 30%-40% of leukemia patients on Dasatinib, and its presence has been associated with increased therapeutic response(12-14).
In contrast to LGL leukemia, the lymphocytosis does not require any intervention, and it ceases once Dasatinib is discontinued, though lymphocyte values can vary during treatment(14).
Similarly, in our patient, lymphocytosis developed two months after Dasatinib treatment, but it decreased to normal values and remained in the normal range after three weeks.
Mustjoki et al.(12), in a comparative study of 55 leukemia patients treated with a TKI, showed that the large granular lymphocytosis (LGL) is a phenomenon that is unique to Dasatinib (present in 38% patients), and not observed in patients who were treated with Imatinib, Nilotinib or Bosutinib, though all TKI’s can lead to clonal expansions of memory cytotoxic T-cells(11).
In patients who developed LGL, both with an underlying diagnosis of CML and Ph+ ALL, molecular responses obtained were superior to the patients who did not have lymphocytosis(13,14). This was seen in our patient, who obtained a complete molecular response by bone marrow PCR about six months after diagnosis, thus after the development of the transient lymphocytosis and LGL.
However, the Dasatinib-induced lymphocytosis has been implicated in the autoimmune-like adverse events reported with this drug, namely lung manifestations (both pleural effusions and lung parenchymal involvement), colitis, and fevers(13).
In a study of 40 patients with CML treated on Dasatinib, 22.5% developed lung abnormalities attributable to Dasatinib; pleural effusion fluid revealed lymphocyte-predominant exudates, and these findings were consistent with pleural biopsy and Bronchoalveolar Lavage fluid (BAL) analyses(8,15). Pleural fluid immunophenotyping done in another trial by Mustjoki et al. in patients treated with Dasatinib who experienced pleural effusions confirmed the presence of a clonal T-LGL cell population(13).
Our patient developed pleural effusions prior to the appearance of lymphocytosis, and temporary discontinuation of the Dasatinib as well as thoracocentesis led to symptomatic improvement of dyspnea.
Another association with the clonal expansion of NK/T-cells in Dasatinib treated patients has been CMV reactivation.
In the Mustjoki et al. trial which included 22 Ph+ ALL patients on Dasatinib with lymphocytosis, 9 patients had modest CMV reactivation, with CMV genome load paralleling the degree of lymphocytosis(13). This resolved without any antiviral treatment, as it did in our patient. Also, CD4+ T-cell counts were noted to be decreased, with an increase in CD8+ T-cells, as seen in our patient(13,18).
Finally, it is important to highlight two more findings related to the Dasatinib-induced LGL process. The first is that it appears that a significant number of patients have clonal lymphocytes present at CML diagnosis, and the clone size expands with the administration of Dasatinib, but not Imatinib, leading to the observed peripheral blood lymphocytosis(20).
In the study population of 34 CML patients (20 of whom subsequently received Dasatinib and 14 were given Imatinib), 83% had a clonal, BCR-ABL1 negative lymphocyte population, whereas clonal lymphocytes were seen in only 8% of healthy control patients(20). In 69% of the cases, the clone found in follow-up samples was identical to the one at diagnosis, supporting the observation that Dasatinib favors the expansion of this preexisting clone.
The second interesting finding comes from a recent study published by Mustjoki et al. in which 55 Ph+ ALL patients on TKI’s had blood samples drawn at specified time points from drug intake(12). It appears that the fluctuation in lymphocyte counts is associated with the timing of blood draws, and the degree of lymphocytosis is dose-dependent, with peak drug and lymphocyte levels at one to two hours after drug intake, and with a rapid decline thereafter.
The authors conclude that the previously observed lymphocytosis is merely a reflection of drug administration-related timing.
In our patient, the LGL lymphocytosis was short-lived, and it did not recur despite frequent blood count monitoring, but Mustjoki’s study does raise the issue of consistent monitoring of lymphocyte counts at a constant time-interval from oral Dasatinib administration. In the future, complete blood counts with differentials drawn precisely one to two hours after Dasatinib should be followed in all patients. Based on the degree of lymphocytosis at that time point, drug dose intensification should be investigated in clinical trials given the favorable clinical outcomes seen in patients with lymphocytosis.