Here we go, let’s just say this is our case;
General Practitioner: I have a 54-year old patient who showed up for her annual, no complaints, feeling fine. But her routine blood work came back with a white count of 44,000, mild anemia and borderline thrombocytosis. She denied any infectious symptoms, and is not taking any steroids or other medications. I gave her a Zithromax prescription and told her to return in two weeks for a repeat CBC, and her WBC count is still high at 51,000. Should I refer her to you?
Hem/Onc MD: I’ve seen a lot of my patients present that way. I’d be surprised if it was an infection. Please have her call my office for an appointment right away. What does the WBC differential show?
GP: Fifty percent neutrophils, 15% monocytes, 4% basophils, and ‘myelocytes and metamyelocytes’ were also noted.
Hem/Onc MD: Any blasts?
GP: Two percent. What do you think?
Hem/Onc MD: Sounds like it’s probably a myeloproliferative disorder, quite possibly CML.
GP: You need me to send along all other annual tests we just did?
Hem/Onc MD:You can but it’s not necessary. We have to do a bone marrow biopsy and aspiration, and I will send the aspirate for flow cytometry and cytogenetics. If she has CML, we’ll be looking to find the ‘Philadelphia chromosome’ present in her metaphases.
Chronic myelogenous leukemia (CML) is a clonal hematopoietic malignancy that is characterized by the overproduction of neutrophils in all stages of maturation secondary to a balanced translocation, t(9;22) (q34;q11.2), between the long arms of chromosomes 9 and 22, which is also known as the Philadelphia or Ph chromosome. The Philadelphia (Ph) chromosome is detected in approximately 95% of CML cases but is cryptic (i.e., not detectable by cytogenetic analysis but rather by the molecular rearrangement present) in all other patients with CML. This translocation creates a unique fusion gene, BCR-ABL, whose protein product, bcr-abl, is a constitutively-active tyrosine kinase that promotes unregulated granulocyte production and in some patients, unregulated platelet production. In the U.S. alone, approximately 4,500 new cases arise annually with a median age of onset of 53 years.
CML has been classified by the World Health Organization (WHO) as a chronic myeloproliferative disorder along with chronic neutrophilic leukemia, chronic eosinophilic leukemia (and the hypereosinophilic syndrome), polycythemia vera, chronic idiopathic myelofibrosis (with extramedullary hematopoiesis), essential thrombocytosis and other unclassifiable chronic myeloproliferative disorders (Table 1.1), because it involves a multipotent hematopoietic progenitor cell; overproduction of one or more of the formed elements of the blood; marrow hypercellularity and megakaryocytic dysplasia; extramedullary hematopoiesis; myelofibrosis and transformation to acute leukemia.
WHO Classification of the Chronic Myeloproliferative Disorders.
- Chronic Myelogenous Leukemia (Ph chromosome, t(9;22)(934;q11) BCR/ABL positive
- Chronic Neutrophilic Leukemia
- Chronic Eosinophilic Leukemia (and the hypereosinophilic syndrome)
- Polycythenia Vera
- Chronic Idiopathic Myelofibrosis (with extramedullary hematopoiesis)
- Essential Thrombocytosis
- Chronic Myeloproliferative Disorder, unclassifiable
However, CML differs from many of the other chronic myeloproliferative disorders because, in contrast to most of them in the absence of effective therapy, its clinical course is usually measured in years rather than decades, terminal transformation to acute leukemia is inevitable and its timing is unpredictable; but most importantly, only CML is associated with the t(9;22) (q34;q11.2) translocation, for which there is now targeted therapy with the drugs imatinib mesylate and dasatinib.
The natural history of CML has been phenotypically divided into three phases: chronic, accelerated and blastic. Eighty-five to 90% of CML cases in the U.S. are diagnosed in the early, or “chronic,” phase of CML. However, patients can present for the first time in any of the three phases, or can transform spontaneously from the chronic phase to blast crisis, even following remission induction. The blastic phase can be myeloid, lymphoid or undifferentiated. The chronic phase has also been divided into early and late stages and as the disease progresses there is an increasing tendency for the development of extramedullary hematopoiesis with enlargement of the liver and spleen; extension of the disease to the lymph nodes is usually a sign of transformation and is associated with the acquisition of additional chromosome abnormalities such as a reduplicated Ph chromosome.
Most descriptions of the natural history of CML derive from an era when medical care was less accessible, blood counts were not routine and the distinction between the various chronic myeloproliferative disorders was largely based on clinical phenotype and conventional cytogenetics. Furthermore, the impact of therapy, particularly with alkylating agents, was often not factored into descriptions of its natural history. Today, with greater access to medical care and the routine use of electronic cell counters, as well as more sensitive techniques for detecting mutations, patients are being identified earlier in the course of their disease and the phenotype of CML, as well as that of the other chronic myeloproliferative disorders, has changed.
Given this increased access to medical care, it is not surprising that a substantial proportion of CML patients are now discovered incidentally during an evaluation for some other medical issue and that close to half are asymptomatic at the time of diagnosis. Fatigue, weight loss and night sweats are common symptoms and appear not to be related to disease stage, though, undoubtedly, in most cases the cause is hypermetabolism caused by the large turnover of myeloid cells. Fever and infection are uncommon. Interestingly, in one series, bleeding was a common feature, which appeared to be related to platelet dysfunction and thrombocytosis; however, in contrast to the other chronic myeloproliferative diseases, thrombosis is not a significant feature of CML. Rapid growth of the spleen can lead to early satiety and splenic infarction, which can be difficult to distinguish from a pulmonary embolism or pneumonia. Splenomegaly is present in over 75% of patients at the time of diagnosis and can be massive.
The hallmark of CML is the autonomous overproduction of myeloid progenitor cells such that the peripheral blood has the histological characteristics of the bone marrow with respect to the presence of circulating myeloblasts, promyelocytes, myelocytes, metamyelocytes, band forms and mature neutrophils that appear to be morphologically normal but lack the expected granules containing leukocyte alkaline phosphatase. The shift in immaturity is such that the number of myelocytes is frequently greater than the number of more mature myeloid cells. The platelet count can be normal or elevated to the extent found in essential thrombocytosis or polycythemia vera. By contrast, erythropoiesis is not increased and anemia is common but usually mild. An increase in circulating basophils is an important characteristic of CML and an increasing basophil count is a harbinger of disease acceleration.
Of course, exceptions to these phenotypic characteristics are well documented. For example, in some patients, thrombocytosis alone, often extreme, is the presenting manifestation of CML, and the characteristic basophilia and low leukocyte alkaline phosphatase score can be absent. In others, there is primarily a mature neutrophilic leukocytosis, making the disease difficult to distinguish phenotypically from the rare disorder, chronic neutrophilic leukemia. In a few patients, both neutrophilia and monocytosis are present, making the disease indistinguishable phenotypically from the myeloproliferative form of chronic myelomonocytic leukemia.
These phenotypic differences are frequently a consequence of the particular site of the breakpoint in the BCR gene. Thus, in most CML patients, the BCR-ABL fusion protein has a molecular mass of 210 kDa (p210) but in patients with a mature neutrophilic leukocytosis, bcr-abl is a 230 kDa protein (p230) and in patients with increased monocytes, BCR-ABL is a185 kDa protein (p190); p190 is also expressed in some patients presenting with acute lymphocytic leukemia.
Eosinophilia can also be a feature of CML and occasionally is sufficiently prominent to pose a problem in distinguishing these patients from those with hypereosinophilic syndrome, chronic eosinophilic leukemia or systemic mastocytosis. Although many of the latter patients respond to imatinib mesylate, the dose employed is lower and their natural history and complications are different.
A low neutrophil leukocyte alkaline phosphatase (LAP) score is a curious feature of CML but one that is not fixed, as the LAP score in CML can increase with infection, myelofibrosis, exposure to corticosteroids and during blast crisis. As mentioned, thrombocytosis, often extreme, can be the sole manifestation of CML and is the one situation in which neither leukocytosis nor a low LAP score may be present. In this situation, given the exquisite sensitivity of RT-PCR (reverse transcriptase-polymerase chain reaction) for BCR-ABL, to avoid false-positive results in the evaluation of a thrombocytosis patient for CML, other tests such as conventional cytogenetics or BCR-ABL fluorescence in situhybridization (FISH) technology should be employed instead for diagnosis.
The bone marrow aspirate in CML is hypercellular with a predominant myeloid hyperplasia with a left shift that will be characteristic of the particular phase of the disease. Thus, the blast cell count is less than 10% in the blood or bone marrow in the chronic phase, from 10% to 19% in the accelerated phase, and >20% in blast crisis. These thresholds represent the guidelines for a recently proposed WHO classification but other definitions exist, with somewhat different criteria (Table 1.2).
Criteria for Accelerated Phase according to MDACC, IBMTR and WHO.
|Blasts + Pro’s
||Unresponsive ↑, persistent ↓
||<100 or >1000 unresponsive
||CE not at diagnosis
||Difficult to control, or doubling <5d
||Megakaryocyte proliferation, fibrosis
NA = Not applicable, CE= Clonal evolution
* Blast phase >20% blasts (>30% for MDACC and IBMTR)
** Basophils + eosinophils
Like the other chronic myeloproliferative disorders, there is megakaryocyte hyperplasia in CML, but in striking contrast to the other disorders, the megakaryocytes are typically small and hypolobulated in CML. Because of the high leukocyte turnover, Gaucher-like cells stuffed with glycolipids may be seen. Myelofibrosis is rarely a presenting feature of CML but can develop with disease acceleration, as can red cell aplasia and lymphadenopathy. Bone infarction and hypercalcemia are usually features of blast crisis.
Given the phenotypic mimicry described above and the advent of targeted therapy, the diagnosis of CML requires the identification of the Ph chromosome or the BCR-ABL fusion gene in those patients in whom expression of the fusion gene is cryptic and conventional cytogenetics is not informative. In this instance, analyses for BCR-ABL by FISH or RT-PCR are diagnostically useful. Conventional cytogenetics has the advantage of identifying genetic abnormalities in other chromosomes besides the Philadelphia chromosome. These abnormalities may have adverse prognostic significance and include trisomy 8, isochromosome 17 and +19. From a therapeutic as well as a diagnostic perspective, a bone marrow aspiration and biopsy, together with conventional cytogenetics and quantitative RT-PCR for BCR-ABL in peripheral blood cells, are the preferred initial studies for CML.
The clinical features at the time of diagnosis have been incorporated into risk scores such as the Sokal and the Hasford. They take clinical features such as age, spleen size, percentage of blasts, basophils and eosinophils into a formula that defines patients with low, intermediate or high-risk disease. These classifications have been useful in defining the probability of progression.
- The Philadelphia chromosome is detected only in the leukemic cells of CML patients and not in any other tissues including normal hematopoietic progenitors. The absence of the Philadelphia chromosome from normal hematopoietic progenitors enables the achievement of a complete cytogenetic response to medical therapy.
- Polymerase chain reaction (PCR) technology can be used to detect BCR-ABL but, strictly speaking, not the Ph chromosome. Along with FISH, PCR can be useful for the diagnosis of CML in the absence of a Ph chromosome. In these cases, the BCR-ABL fusion is the result of a cryptic chromosome rearrangement. While a subset of normal healthy volunteers can be shown to harbor BCR-ABL, when sensitive methods of detection such as PCR are used, the Ph chromosome is never observed in healthy volunteers. Occasionally, cases of chronic neutrophilic leukemia are associated with a variant form of BCR-ABL. The reciprocal translocation, t(15;17), is associated with acute promyelocytic leukemia and results in the fusion of the PML and RAR genes.
CML Cytogenetics and FISH. Not the best picture but I'm not an artist. Where is Jess when you need her? :)
Pathogenesis of CML
In murine model systems, BCR-ABL is capable of causing a CML-like myeloproliferative syndrome. This phenotype is dependent upon the tyrosine kinase activity of BCR-ABL In human CML, the BCR-ABL-selective kinase inhibitor, imatinib mesylate, leads to a high response rate, which validates the importance of the tyrosine kinase activity of BCR-ABL in the pathogenesis of the disease. Preclinical studies suggest that SRC family kinases may play a role in CML disease progression to lymphoid blast phase. Although epigenetic changes in gene expression may play a role in the pathogenesis of CML, reversal of these changes does not appear to result in a significant percentage of remissions. There is no evidence that TNF-a activity plays a significant role in the pathogenesis of CML. While the normal ABL gene has a DNA binding domain that is retained in BCR-ABL, BCR-ABL does not localize to the nucleus and is not known to have appreciable DNA binding activity.
The molecular pathogenesis of CML has been the subject of intense scientific investigation. The Philadelphia (Ph) chromosome is a reciprocal translocation between chromosomes 9 and 22 that results in the formation of a fusion BCR-ABL gene, which encodes a protein tyrosine kinase with dysregulated enzymatic activity. The tyrosine kinase activity of BCR-ABL is essential to its ability to generate a CML-like state in lethally-irradiated mice transplanted with BCR-ABL-transduced stem cells. BCR sequences also play a vital role in facilitating oligomerization of BCR-ABL, which leads to activation via trans-phosphorylation.
At the molecular level, different BCR-ABL transcripts can be generated, depending upon the precise location of the BCR gene breakpoint. Most commonly, a 210-kd protein (P210) is formed. Occasionally, most often in cases of Philadelphia chromosome-associated acute lymphoblastic leukemia (ALL), a P190 is detected. Rarely, particularly in cases of chronic neutrophilic leukemia, a P230 isoform is created.
Because of differences in the chromosome 22 breakpoints, Ph+ ALL is frequently associated with a 190 kD protein (p190), whereas CML is most often associated with a 210 kD form (p210).
In vitro studies demonstrate that the p190 isoform of BCR-ABL harbors significantly more kinase activity than p210. This suggests some degree of negative regulation of the BCR-ABL p210 kinase by sequences in BCR that are absent in BCR-ABL p190. In addition to an increased level of ABL tyrosine activity as a result of the BCR-ABL fusion, BCR sequences contain sequences important for oligomerization, which are essential for the ability of BCR-ABL to induce cellular transformation in vitro.
A Case Study
A 47-year-old salesperson is diagnosed with chronic myeloid leukemia when a routine blood test revealed an elevated WBC of 74.3 x109/L with 4% basophils, 12% metamyelocytes, 6% myelocytes and 5% blasts. The platelet count was 563 x109/L and the hemoglobin 13.1 g/dL. The spleen was palpable 4 cm below the costal margin. The Sokal risk score is low. (To calculate Sokal risk score: http://www.roc.se/Sokal.asp)
The treatment of CML has changed significantly in recent years. For many years, hydroxyurea and busulfan were used as primary therapy. Although they effectively controlled peripheral blood counts in most patients, they rarely if ever produced cytogenetic responses. More recently, interferon alpha (IFN-α) became the standard therapy for most patients. The use of IFN-α resulted in cytogenetic responses in 40-50% of patients and they were complete in 5-20%. The addition of ara-C to IFN-α resulted in improved response rates, with some studies suggesting a survival advantage compared to IFN-α alone. Treatment with IFN-α, however, was associated with significant toxicity.
Imatinib, at a standard dose of 400 mg daily, has resulted in a complete cytogenetic response rate of 69% after 1 year of therapy and with five years of follow-up, 87% of patients have achieved this level of response at some time during imatinib therapy. A randomized trial (the International Randomized Study of Interferon and STI571 “IRIS” trial ), comparing imatinib to IFN-α + ara-C, demonstrated a significant advantage with imatinib in response rate, toxicity profile and progression-free survival. These responses have been largely durable. The estimated survival free of transformation to accelerated and blast phase CML at 5 years is 93% and the overall survival on an intent-to-treat analysis was 89% . Indeed, the rate of progression has decreased in the last 2 years of follow-up with <1% of patients losing their response or progressing in the 5th year of follow-up. Thus, imatinib has become the standard therapy for patients with CML.
The treatment of choice today for most patients with CML is imatinib and the standard dose is 400 mg daily. The role of hydroxyurea is merely to control the WBC for some time, for example, while determining if a patient has the Philadelphia chromosome or not. It should not be considered definitive therapy. The IRIS trial clearly documented the benefit of imatinib over the combination of interferon and ara-C in terms of response and tolerability. A survival advantage was not documented, probably because of the crossover design of the study, where patients not responding or tolerating their initial therapy were allowed to receive the alternative therapy. Because of this, over 90% of patients initially receiving IFN-α rapidly switched to receive imatinib. However, recent studies comparing imatinib to historical series of patients treated with interferon-based therapy have shown the expected survival advantage.
Several single-arm studies have suggested that higher doses of imatinib may be superior in terms of cytogenetic and molecular responses, and some have suggested a benefit in progression-free and transformation-free survival compared to historical controls. On-going randomized studies are trying to confirm the enhanced efficacy of higher doses of imatinib. Until these studies confirm an advantage, the standard dose remains 400 mg daily. There are no studies to show any advantage of imatinib in combination with interferon or any other agents. Although transplant has been used for many years, imatinib is considered today the best choice for most adults with CML. With a transplant-related mortality during the first year of at least 15% to 20% (and increasing with age) and the very favorable results with imatinib (survival rates of 92% at 5 years), almost all current patients should be considered for a trial with imatinib as first-line treatment.
For many years, interferon alpha was the standard therapy for patients with CML in the chronic phase who were not candidates for a stem cell transplant. The rate of complete cytogenetic response with interferon alpha varied in different studies (between 5% and 25%). Higher response rates were achieved with the combination of interferon alpha and cytarabine — up to 30% to 40%. Two randomized trials demonstrated this improved outcome with this combination that extended to progression-free survival, and improved survival was demonstrated in one of these trials.
It is important to educate the patient as to how to take imatinib. Although the initial studies were done with imatinib administered on an empty stomach, the current recommendation is to take it with a meal, preferably a light one. This has been shown to decrease the frequency of gastrointestinal adverse events, particularly nausea. While receiving imatinib, the patient should avoid acetaminophen because of the potential risk of liver toxicity. It is also recommended that grapefruit be avoided because of its potential interaction with the metabolism of imatinib. Other medications metabolized through the P450 can be administered but this should be done with caution. For example, patients who receive oral anticoagulants, (e.g., warfarin), should be closely monitored when imatinib is first administered and whenever a change in dose is introduced. For patients receiving a daily dose of 800 mg daily, splitting the dose into two equal doses of 400 mg is acceptable and may lead to improved tolerance.
Imatinib can be administered to the patient as soon as the diagnosis of CML is confirmed. It is not necessary to give other therapies prior to the start in an attempt to bring the WBC down.
Patients treated with imatinib should start therapy with the optimal dose at the outset of therapy. There is no evidence that a slow dose escalation improves, for example, tolerance. Indeed, it has been suggested that the use of suboptimal doses of imatinib may create conditions that increase the risk of developing resistance. In addition, there is growing evidence that the earlier the patient achieves a profound reduction in tumor load, the better their long-term outcome. Thus, the standard dose for patients receiving imatinib should be given at the start of therapy.
Allogeneic Bone Narrow Transplantation
Allogeneic bone marrow transplantation (BMT) can be curative in a significant proportion of CML patients eligible to receive this approach. Survival rates of 60-80% are frequently achieved in centers with experience. Besides availability of a donor and age, the major consideration is the risk of early mortality and long-term complications. Early mortality is usually lowest among younger patients, particularly those younger than age 20 to 30 years. The occurrence of late relapses, (more than 5 years after transplant), are increasingly being recognized and should also be taken into consideration. Extensive chronic graft-versus-host disease can be seen in as many as 60% of patients treated and other late complications may occur.
All of these issues need to be evaluated before opting for bone marrow transplantation. In addition, the experience level at the site where the transplant is to be performed needs to be considered. Reduced intensity conditioning regimens have expanded the population that may be eligible for transplant to include older patients, but the risk of relapse may be higher than with standard therapy. Transplant from alternative donors, (i.e., unrelated, cord blood), are also being evaluated but the potential for increased risks and lack of long-term data should be taken into account. Matched unrelated transplants may be associated with increased mortality and graft-versus-host disease, although molecular matching has decreased these risks considerably.
“Stem cell cures patients with CML”?
This has been the most common statement in the CML literature for many years and is still frequently seen today. Undoubtedly, stem cell transplant has cured many patients with CML. However, late relapses after transplant, (i.e., occurring more than 5 years after transplant), are increasingly being reported. The risk is small but constant and present as late as 20 years after transplant, (longer follow-up data is not available), complicating the definition of “cure” after transplant.
Furthermore, a subset of patients treated with interferon achieved undetectable transcript levels, (i.e., “complete molecular remission”), and none of these patients had relapsed after 10 years of follow-up. Studies following patients long-term after discontinuation of therapy with interferon have shown that approximately 50% had not relapsed. Thus, it is clear that some patients have been cured with interferon, although these represent a small subset of patients.
Stem cell transplant (SCT) is still a valuable treatment option for patients with CML. The best results are achieved in patients transplanted in the first chronic phase compared to those transplanted in the second chronic phase, (i.e., returning to the chronic phase after therapy for CML in the accelerated or blast phase). Results are also better for patients transplanted with cells from an HLA-identical sibling donor. Before the imatinib era, some reports initially suggested that prior therapy with interferon alpha adversely affected the outcome after stem cell transplant. However, multiple subsequent reports demonstrated this was not the case. Some of these subsequent reports suggested that suspending interferon at least 3 months prior to transplant could correct this alleged adverse influence. Studies reporting a similar analysis with prior exposure to imatinib have shown that prior exposure to imatinib does not adversely affect the outcome after SCT.
Frequently, after SCT, patients still have positive PCR for the first few months. However, only when these remain positive more than 6 months after transplant is there a correlation with a higher risk of relapse. Intervention for positive PCR during the first 6 months is usually not indicated.
Although the highest risk of relapse after transplant is in the first 2-3 years, it has become clear that late relapses do occur. There is a small but constant risk of relapse after 5 years, extending as far as 20 years after transplant. In addition, late complications of transplant include chronic graft-versus-host disease, osteoporosis, cataracts, secondary malignancies and infertility, although most patients who survive long-term after transplant have an adequate quality of life.
Cancer Treatment = Complications
*** After 1 month of therapy, the white cell count has come down to 2.8 x109/L with 43% neutrophils and ANC [(absolute neutrophil count) 1.2 x109/L]. At this time, your recommendation should be to continue therapy unchanged and check CBC in 1 week.
The current recommendation for patients with CML in the chronic phase treated with imatinib regarding the management of myelosuppression is to hold therapy if the patient develops neutropenia or thrombocytopenia grade 3 or 4. Neutropenia is defined herein as an absolute neutrophil count of less than 1 x109/L. Some studies have actually used a lower threshold, (ANC 0.5 x109/L, which represents grade 4 neutropenia), as most patients recover rapidly and septic episodes are extremely uncommon. The treatment is reinstituted once the neutrophil count recovers above these levels. It is not necessary to discontinue therapy or to decrease the dose at absolute neutrophil counts higher than this level.
Although filgrastim has been shown to be of benefit in some patients with neutropenia, it is usually not recommended at the first occurrence of neutropenia, as this adverse event is relatively common during the first 2-3 months and because after a brief interruption of therapy most patients recover rapidly and do not experience this problem again. Filgrastim is best reserved for patients who experience prolonged or recurrent episodes of grade 3 or 4 neutropenia that compromise adequate dosing of imatinib. After treatment interruption for grade 3 neutropenia, treatment can be re-started at the same dose, provided the neutrophil counts recover within 2 weeks. If recovery takes longer than 2 weeks, dosage should be reduced to 300 mg if the patient was receiving therapy with 400 mg daily, (to 600 mg if the patient was receiving 800 mg daily and to 400 mg if previously receiving 600 mg).
One of the most common adverse events associated with imatinib therapy is myelosuppression. Myelosuppression is more common in patients who have failed prior IFN-α and is dose-related. Myelosuppression occurs most frequently during the first 2-3 months of therapy. After this time, few patients develop significant myelosuppression and most can continue therapy uninterrupted.
Overall, among patients treated with imatinib as first-line therapy, grade 3 or 4 neutropenia has been reported in 17%, thrombocytopenia in 9% and anemia in 4%. However, only 4% of patients develop neutropenia after 2 years of therapy, while 2% have thrombocytopenia and 2% become anemic. After 4 years of therapy <1% demonstrate any of these complications. It is very uncommon to have patients develop infectious or hemorrhagic complications associated with these events. In fact, in most instances myelosuppression is transient and self-limited and can be managed with short treatment interruptions. The use of hematopoietic growth factors, (filgrastim for neutropenia, oprelvekin for thrombocytopenia, erythropoietin or darbepoetin for anemia), has been reported to be effective in patients with recurrent or persistent cytopenias.
As with neutropenia, the recommendation for thrombocytopenia is to interrupt therapy for grade 3 or higher toxicity. Grade 3 thrombocytopenia is defined as a platelet count below 50 x109/L. The rules for re-instituting therapy are similar to those for neutropenia — if recovery above these levels occurs within 2 weeks, therapy is re-instituted at the same dose; if recovery takes longer than 2 weeks, a dose reduction, similar to the approach for neutropenia, is recommended.
Several studies have shown that the use of eythropoietic growth factors can result in improved hemoglobin levels in the majority of patients treated with imatinib who develop anemia. However, it should be emphasized that this is not an approved use of erythropoietin or darbepoetin at the moment. These agents are approved for use in patients with anemia associated with cancer therapy in non-myeloid malignancies.
*** A patient develops a rash that covers approximately 60% of the arms and the upper chest. There are a few small papules in the back and thighs. The rash is pruritic. The patient is taking 400 mg daily of imatinib. You continue therapy with imatinib at the same dose and treat the rash.
Occasionally, patients may develop skin rash while receiving therapy with imatinib. Prompt intervention is important to minimize treatment interruptions and to conserve dose intensity. Dermatologic treatment should be initiated at the first evidence of rash that may be related to imatinib. If the rash is grade 1 or 2, therapy with imatinib can continue but the rash should be treated. Patients may receive symptomatic therapy with antihistamines for the pruritis and topical and/or systemic steroids as needed. This controls the rash in most instances.
The patient described above has a grade 2 rash, because it is symptomatic, (thus not grade 1), and covers less than 50% of the total body surface, (thus not grade 3). If the rash does not respond to adequate management or progresses to grade 3, then imatinib therapy should be interrupted until it resolves to at least a grade 1. Therapy is then resumed at a lower dose: 300 mg daily if the patient was receiving therapy with 400 mg daily, 400 mg if the patient was receiving 600 mg daily and 600 mg if the patient was receiving 800 mg.
*** A patient with CML who is receiving therapy with imatinib 400 mg daily develops elevation of aspartate transaminase (AST) to 2-3 times the upper limit of normal. You continue therapy unchanged.
Proper monitoring and intervention to manage adverse events are important to optimize therapy in a way that preserves patient safety and offers the best possible outcome. In general, treatment interruptions are recommended when patients develop grade 3 or higher non-hematologic toxicity. For liver toxicity, this is defined as an elevation of transaminases to >5-20 times the upper limit of normal (ULN). Thus, the patient described above has grade 2 liver toxicity. Continuation of therapy unchanged is indicated in this instance. However, this should be accompanied by investigation of other potential causes of liver damage, (e.g., the concomitant use of other drugs that might cause liver damage), and close monitoring. The patient’s liver function tests should be monitored with increased frequency until the event is resolved or has stabilized.
If grade 2 toxicity persists, a temporary treatment interruption may be indicated and dose reduction considered. If the patient develops grade 3 liver toxicity, these measures should be instituted immediately with therapy restarted at a lower dose once the enzymes have returned to at least grade 1 level (i.e., <2.5 ULN). Sometimes patients require frequent interruptions because of liver toxicity. Although an occasional patient will eventually not show evidence of liver toxicity after multiple treatment interruptions, a patient with persistent and/or recurrent liver toxicity should discontinue therapy permanently and be offered alternative therapy with dasatinib.
Imatinib and Cardiac Failure
A recent report in the literature and, unfortunately, in the media suggested there might be an increased risk of cardiac toxicity and heart failure for patients receiving imatinib. Unfortunately, the reports about this possible adverse event misrepresented a very valuable study, mostly about potential cardiac effects of imatinib in animal models. In that study, a few (10) patients who had developed congestive heart failure while receiving therapy with imatinib were also presented. Although congestive heart failure has been described occasionally among patients with CML treated with imatinib, the incidence appears to be very low. In addition, when it has been reported, it is most frequently seen in older patients and in patients with pre-existing cardiovascular problems, (e.g., cardiomyopathy from interferon therapy, hypertension, coronary artery disease etc.), or risk factors for heart failure. (e.g., diabetes, prior use of cardiotoxic agents such as anthracyclines, etc.).
In the overall universe of patients treated with imatinib, the incidence of congestive heart failure does not appear to be any higher than would be expected in a population of individuals of similar age. Still, as for all other adverse events that may occur with imatinib, adequate monitoring and intervention are required. Imatinib may result in fluid retention and opportune therapy with diuretics when indicated is usually effective. For the occasional patient with symptoms that suggest congestive heart failure, an adequate work-up is indicated. Routine vigilance with echocardiograms, EKG or other similar cardiac tests is not indicated for the great majority of patients.
Treatment Failed?: Cytogenetic response
*** After 3 months of therapy with imatinib 400 mg daily, a patient with CML who started therapy in the chronic phase has achieved a complete hematologic response — a routine karyotype analysis in 20 metaphases shows 1 with the Philadelphia chromosome and 19 diploid.
Patients who have not achieved a complete cytogenetic response after 3 months of therapy still have over a 50% probability of achieving a complete cytogenetic response at 24 months with continuation of therapy. If no cytogenetic response is achieved by 6 months, the probability of achieving a complete cytogenetic response with continuation of therapy is approximately 15% compared to 50-80% if at least a minor cytogenetic response has been achieved. Therefore, no cytogenetic response after 6 months is considered as failure to imatinib.
By 12 months, patients with a partial cytogenetic response still have nearly a 50% probability of achieving a complete cytogenetic response with continuation of therapy compared to <20% for those with less than a partial cytogenetic response. However, earlier responses correlate with an improved long-term outcome. For example, patients who have not achieved a complete cytogenetic response after 12 months of therapy have a significantly lower probability of being alive and free from transformation or loss of response at 54 months (72%) compared to those with a complete cytogenetic remission (>90%).
This information has been incorporated into the current expert recommendations for the management of patients with CML. According to these guidelines, not achieving a complete cytogenetic response at 12 months from the start of therapy is considered a suboptimal response to therapy, whereas not achieving a complete cytogenetic response after 18 months of therapy is considered failure to therapy.
The standard definitions of cytogenetic response are based on routine cytogenetic analysis (karyotype) of at least 20 metaphases. A complete cytogenetic response is achieved when none of the 20 metaphases has the Philadelphia chromosome present following therapy. If the Philadelphia chromosome is present in 1-34% of metaphases, this constitutes a partial cytogenetic response. A minor cytogenetic response represents the presence of Ph in 35% to 95% metaphases. Some recent reports have subdivided minor responses into minor (35-65% positive) and minimal (>65% positive), although there does not appear to be any difference in the long-term outcome of patients between these two groups.
Based on the available data regarding the optimal responses that correlate with improved event-free survival, the definitions of failure to imatinib and suboptimal response have been established. These definitions take into consideration not only the response achieved but also the time at which responses are achieved.
Definitions for Failure and Suboptimal Response to Imatinib
||No MMR (<3-log BCR-ABL/ABL)
||Loss of CHR
Loss of CCgR
Loss of MMR
HR: Hematological remission; CHR: Complete hematological remission; Ph+ve: Philadelphia chromosome positive; MMR: Major molecular response; CCgR: Complete cytogenetic response; CE: Clonal evolution.
*** A patient with CML in the chronic phase has been receiving imatinib at a dose of 400 mg daily with adequate tolerance. After 12 months of therapy, the patient has achieved only a partial cytogenetic remission, (i.e., 1-34% Ph+ metaphases).
In the patient presented, a partial cytogenetic response has been achieved at the 12-month mark, (a suboptimal response). This patient still has approximately an 80% probability of achieving a complete cytogenetic response with continuation of therapy and their probability of survival free from transformation is 93%. The response cannot be considered failure and consideration of stem cell transplant with the involved risks is not warranted; similarly, the response does not meet the current requirements for the use of dasatinib. However, it should be considered a suboptimal response, as a complete cytogenetic response is associated with a survival free from transformation at 5 years of 97%.
Increasing the dose to optimize response is recommended for all patients with suboptimal response to imatinib. All published data with dose escalation have been based on doubling the dose the patient is receiving. For example, this patient would receive 800 mg daily. Ongoing studies are evaluating the role of dasatinib and nilotinib in this setting, (suboptimal responses). Until these data are available, the standard in this situation should be dose escalation.
A complete cytogenetic response has been associated with improved survival. The presence of chromosomal abnormalities in other chromosomes in addition to the Philadelphia chromosome can be detected by a chromosome analysis (karyotype) and this phenomenon is associated with a worse outcome. Historical data from patients treated with interferon alpha have demonstrated that achieving a complete cytogenetic response correlates with an improved probability of survival. On a long-term follow-up, 78% of patients who achieved this response were alive after 10 years, compared to only 39% of those achieving a partial cytogenetic response and 25% of those with lesser responses.
There is usually a good correlation between the results obtained with cytogenetic analysis and FISH, although FISH has some false positivity. False positivity is low with the modern probes used, (usually at approximately 1-1.5%), but one has to be aware of the level of false positivity to be able to interpret positive results at low levels. However, molecular response is assessed by real-time PCR, which is 3-4 logs more sensitive than FISH and cytogenetic analysis.
The definition for major molecular remission introduced in the IRIS trial is a 3-log reduction of transcript levels compared to a standardized baseline established in untreated patients. In this study, the ratio of BCR-ABL to BCR was assessed in a subset of patients prior to the start of therapy. The average of the results obtained from these patients was established as the baseline value from which patients would have to achieve a 3-log reduction to be considered major molecular remission. Patients who achieve a major molecular remission after 12 months of therapy have an improved probability of survival free of progression and survival free from transformation, compared to those without a major molecular response. A FISH analysis identifies cytogenetic response, not molecular response. PCR is at least 3-4 logs more sensitive than the standard FISH analysis.
Duration of Therapy
The current recommendation regarding duration of therapy with imatinib is to continue therapy indefinitely. There are a few small series of patients who have achieved a complete molecular remission, (sustained for more than 1 year), who stopped therapy with imatinib. Although in some patients no relapse had occurred at the time of the report, in most instances a relapse was seen, frequently as early as 3 months after the treatment was discontinued. Thus, for the moment, it is not recommended to stop therapy even after achieving a molecular response. Studies looking at strategies that may allow safe treatment interruption are being conducted.
Although PCR is a very sensitive technique for detection of the BCR-ABL chimeric gene, it still has a limit of detection. Even if PCR becomes negative, residual disease below the level of detection of the test might still be present. It is probably more appropriate to characterize this result as “undetectable transcript” rather than as “complete molecular remission.”
All done. All of a sudden imatinib doesn’t work anymore so you double the dose but still nothing.
Do we have any other options? Yes.
Imatinib at 400 mg/day is highly effective for the treatment of newly diagnosed chronic phase CML, achieving complete cytogenetic remissions in approximately 70% of patients after twelve months of therapy. Data from phase III studies reveal that survival without CML-related death in imatinib-treated patients exceeds 95% with five years of follow-up. Historically, the median survival of chronic phase CML patients was approximately 5-7 years. Imatinib is clearly providing a survival advantage in chronic phase CML patients.
Although complete cytogenetic remissions are common in CML patients treated with imatinib, the vast majority of patients have evidence of residual disease when sensitive methods of detection such as PCR are utilized. Additionally, most patients who are negative by PCR will develop detectable disease within 12 months of treatment interruption.
Although the progression of CML involves the accumulation of mutations that presumably cooperate with BCR-ABL to generate a true acute leukemia phenotype, initial response rates to targeted agents are, nonetheless, substantial. The durability of these responses is typically limited and, therefore, consolidative approaches such as allogeneic stem cell transplantation are recommended when available.
Although survival rates in imatinib-treated chronic phase patients are high with five years of follow-up, approximately 15% of patients have lost either an established hematologic or cytogenetic response, or have progressed to accelerated or blast phase. These patients are clearly in need of alternative therapies such as dasatinib.
Loss of response to imatinib is nearly always the result of loss of BCR-ABL inhibition. The best defined mechanisms are the acquisition of kinase domain mutations within BCR-ABL and overexpression of BCR-ABL through either genomic amplification or acquisition of additional Ph chromosomes.
Imatinib is metabolized in the liver by CYP3A4 and it is possible that CYP3A4 inducers may decrease exposure to imatinib but this has not been convincingly demonstrated in patients. SRC activation has been demonstrated in a small number of cell lines that were established from imatinib-resistant patients but its contribution to clinical imatinib resistance remains to be defined. Interestingly, mutations in two other imatinib targets, KIT and PDGFR, have been detected in cases of disease states, (e.g., GIST and hypereosinophilic syndrome), driven by these genes that have lost response to imatinib. However, studies of imatinib-resistant CML have failed to identify resistant mutations in kinases other than BCR-ABL. Of note, cases of non-small lung cancer that have lost an established response to gefitinib or erlotinib are most frequently associated with the acquisition of kinase domain mutations in EGFR.
Molecular Mechanisms of Acquired Imatinib Resistance
Although a trial of dasatinib or nilotinib is likely to be effective for patients who have lost a response to imatinib, it is important to determine if patients harbor the T315I mutation, which fails to respond to these agents. If so, proceeding directly to BMT or pursuing third generation kinase inhibitor therapy, (e.g., agents designed to specifically target the T315I), in a clinical trial setting are indicated.
Although dasatinib is a potent SRC inhibitor, nilotinib does not inhibit SRC family kinases. Both drugs appear to be efficacious for the treatment of imatinib-resistant CML. The activity of the drugs against nearly all imatinib-resistant mutations in vitro is responsible for their clinical utility in imatinib-resistant CML. Their increased potency is valuable in cases of imatinib resistance driven by overexpression of BCR-ABL and may lead to deeper remissions in imatinib-naïve cases. Both dasatinib and nilotinib will be compared with imatinib in newly diagnosed cases of CML in large clinical trials.
While both dasatinib and nilotinib have activity against most imatinib-resistant mutations in vitro, one imatinib-resistant mutation, BCR-ABL/T315I, is highly resistant to both these agents and is likely to be responsible for most cases of loss of response to these agents. Also, it is likely that a small number of additional mutations will contribute to resistance to these agents. To date, no responses have been observed in any patient who harbored this mutation prior to therapy with either dasatinib or nilotinib. It is, therefore, reasonable to exclude such patients from being treated with these agents.
Beyond kinase domain mutations, it is expected that a small number of cases will be associated with activation of alternative growth promoting pathways. Although BCR-ABL overexpression may be encountered in cases that develop resistance to second generation agents, the increased potency of dasatinib and nilotinib will likely minimize the contribution of this mechanism of loss of response to imatinib in cases treated with these inhibitors.
*** A patient has been receiving imatinib for 3 years. He achieved a complete cytogenetic remission after 12 months of therapy but has never achieved a major molecular remission. He has been receiving therapy with imatinib 400 mg daily. On his routine 3-month follow-up, he is found to have 16/20 (80%) metaphases Philadelphia chromosome positive. His peripheral blood is still compatible with complete hematologic remission and there is no splenomegaly. He has been tolerating imatinib well. A mutation analysis shows a mutation E255K (P-loop mutation). The patient has no sibling donors. Now what?
The loss of a complete cytogenetic remission while on adequate therapy with imatinib is considered failure to therapy by current guidelines. Five-year follow-up of patients with chronic phase CML who received imatinib as front-line medical therapy suggests that approximately 17% of patients have either lost an established response, (hematologic or cytogenetic), or have progressed to accelerated or blast phase. Additionally, nearly 10% of patients never achieve a major cytogenetic response and approximately 2-5% of patients are unable to tolerate therapeutic doses of imatinib (>300 mg daily).
The best results with interferon are achieved when it is used in the first year after the diagnosis of CML; after that, the rate of major cytogenetic response drops significantly to less than 10% probability. There is only anecdotal experience with the addition of interferon alpha to imatinib after the patient fails imatinib — the results have not shown any significant benefit and tolerance has not been good. Discontinuing imatinib and starting interferon alpha have resulted in no benefit and toxicity is high. Thus, using interferon, whether alone or in combination with imatinib, would not be an adequate choice. Hydroxyurea can only control the WBC but has no benefit in terms of cytogenetic response, which is what this patient needs. A stem cell transplant can be considered in this patient but, in the absence of a sibling donor, the risk increases. It may be adequate to start a search for a donor in the bone marrow registry.
However, with the recent introduction of the new generation of tyrosine kinase inhibitors, these targeted therapies have become the treatment of choice for most patients. Dasatinib has induced major cytogenetic responses in 51% of patients (complete in 40%). Response rates are nearly identical whether a mutation is present or not. According to the available follow-up data, responses appear durable, with approximately 90% sustained for more than 1 year. Similar responses have been reported with nilotinib, although this agent is currently not yet FDA approved. Increasing the dose of imatinib has been used and before the introduction of dasatinib this would have been the treatment of choice. However, a randomized trial that compared an increased dose of imatinib (800 mg) to starting dasatinib in patients who failed imatinib at doses of 400 to 600 mg daily showed a significantly higher response rate and improved time to progression with dasatinib.
- Dasatinib has been approved by the U.S. F.D.A. for imatinib-resistant and imatinib-intolerant CML, as well as for Philadelphia chromosome-positive ALL cases that are resistant to or intolerant of prior therapy. After five years of imatinib therapy, approximately 20-30% of chronic phase CML patients are eligible for dasatinib therapy. Studies to evaluate the efficacy of dasatinib in newly diagnosed CML patients are ongoing. Although dasatinib harbors activity against the KIT tyrosine kinase, it is not presently indicated in imatinib-resistant cases of GIST.
Doses below 300 mg daily are generally considered suboptimal. For example, in the original phase I study of imatinib after interferon failure, patients treated at doses below 300 mg had a significantly lower response rate compared to those treated at doses of 300 mg or higher,(hematologic response 53% vs. 98%, cytogenetic response 11% vs. 54%, respectively).
Patients with >35% Ph+ metaphases after 6 months of therapy are considered to have a suboptimal response and dose escalation is usually recommended. Some patients are not be able to tolerate dose escalation. Continuing imatinib therapy, even at the same dose, with sustained corticosteroid therapy can be risky. It is better, therefore, to switch these patients to dasatinib. The current studies of new generation tyrosine kinase inhibitors (nilotinib, dasatinib) show very little cross-toxicity. Thus, it is unlikely that these patients will develop skin toxicity with dasatinib and a more sustained therapy could lead to an improved response.
Although side effects are observed with dasatinib, patients who were unable to tolerate imatinib because of grade 3 or 4 non-hematologic toxicity generally did not have recurrent toxicity on dasatinib. Although both drugs can cause bothersome side effects, there does not appear to be significant overlap of symptoms in patients who have been treated with both agents.
Although these agents are still young and little data exist in this regard, a recent publication has suggested that approximately 50% of patients who have failed imatinib and nilotinib can respond to dasatinib, although not always with a cytogenetic response. Preliminary observations with nilotinib suggest that the opposite may also be true in some patients, (i.e., response to nilotinib after failure to imatinib and dasatinib). This is obviously very important because it offers the patient the possibility of effective second- and third-line therapies.
Dasatinib is generally very well tolerated but some adverse events can be observed. Fortunately, most of these are mild and can be managed with timely intervention. Headache has been reported in approximately 30% of the patients. However, headache is rarely severe and very seldom requires treatment interruptions or dose reductions. It responds to analgesic therapy and frequently disappears spontaneously despite continuation of therapy.
Myelosuppression can occur in patients treated with dasatinib. Grade 3-4 neutropenia or thrombocytopenia can be seen in up to 45% of patients treated in the chronic phase. The incidence may be higher in patients treated in the advanced stages of the disease but in these instances it is frequently difficult to differentiate from the cytopenias that are characteristic of the disease itself. The myelosuppression is most frequently transient and does not always require dose reductions.
Pleural effusions can be seen in a small subset of patients treated with dasatinib. Among patients treated in the chronic phase, the incidence has been reported at 17%, but only 3% are grade 3 or 4. Early identification is important as many patients will respond to treatment with diuretics and/or a short course of systemic corticosteroids. Pleural effusion is more common in the more advanced stages of the disease, reaching approximately 30% in the blast phase, but it is usually grade 1 or 2. Overall, only 8% of patients with CML in the chronic phase treated with dasatinib have been reported to have discontinued therapy because of adverse events.
Dasatinib has substantially greater potency against BCR-ABL than imatinib. BCR-ABL-positive hematopoiesis is typically dramatically inhibited, which can lead to cytopenias in a substantial proportion of patients. Therefore, careful monitoring of blood counts is indicated. Cytopenias should be managed with growth factors and transfusions as indicated.
If the patient harbors a substantial number of BCR-ABL-negative hematopoietic progenitors, cytopenias typically resolve and the patient is unlikely to develop recurrent cytopenias. Because no grade 3 or 4 cytopenias have been observed when therapeutic doses of dasatinib have been administered to solid tumor patients, it is not likely that the drug is inherently myelosuppressive.
Accelerated Phase of CML
The accelerated phase of CML is defined by WHO as the presence in the or marrow of 10-19% blast cells and 20% or more blood basophils, persistent thrombocytopenia (<100,000/μL) unrelated to therapy blood or thrombocytosis (>1,000,000/μL) unresponsive to therapy, together with increasing splenomegaly and leukocytosis unresponsive to therapy and cytogenetic evidence of clonal evolution. Other definitions specify a blast cell count of 15-29% in the blood or marrow; greater than 30% blast cells plus promyelocytes together in the blood; or bone marrow with blast cells less than 30%.
*** For the past two days, the patient has become progressively short of breath with a nonproductive cough and chest pain on exertion. On physical examination, he is acutely dyspneic and diaphoretic. The pulse is 110 per minute and the blood pressure is 140/80. Funduscopic examination reveals scattered retinal hemorrhages. There is jugular venous distention and breathing is labored. There is bilateral axillary adenopathy. The chest is clear to percussion and auscultation but there is sternal tenderness. Cardiac examination reveals a soft systolic murmur and an accentuated pulmonary second sound. Abdominal examination reveals hepatosplenomegaly with tenderness laterally over the spleen but no rub is heard. Examination of the extremities reveals pretibial edema and scattered ecchymoses. An ECG shows sinus tachycardia and right axis deviation and a chest x-ray demonstrates increased interstitial markings. On a chest spiral CT, there are thickened interlobar septa with patchy areas of ground glass opacity without focal consolidation. An arterial blood gas reveals a pO2 of 55 mm Hg on room air. The uric acid is 9 mg %, the serum creatinine is 1.2 mg % and the electrolyte panel is normal except for a mild respiratory alkalosis.
This is called Pulmonary hypertension secondary to pulmonary leukostasis.
The clinical picture is most compatible with pulmonary leukostasis in a patient with an elevated leukocyte count and an increased proportion of blast cells. Exertional angina and fluid retention are concomitants of pulmonary hypertension that can occur without the presence of coronary artery disease. Dyspnea in this situation is a consequence of hypoxia. The spiral CT makes it unlikely that a pulmonary embolism is the cause of the right heart strain. Tumor lysis syndrome is characteristically seen following chemotherapy-induced tumor reduction. It is manifested by hyperphosphatemia, hypocalcemia, hyperkalemia, hyperuricemia (>10 mg %) and renal failure.
The apparent failure of the supplemental oxygen therapy is a test-tube artifact resulting from increased oxygen consumption by the high number of leukocytes. Although pulse oximetry is subject to technical inaccuracy, in this instance it would be a useful measure of in vivo blood oxygenation. The apparent hypoglycemia also occurs because of the consumption of glucose by the large number of leukocytes in the blood sample and requires no therapy. Ice alone is often not sufficient to prevent consumption of oxygen or glucose in a blood sample containing a large mass of leukocytes, and the addition of a metabolic inhibitor such as sodium azide is usually necessary for this purpose.
Treatment for Pulmonary hypertension secondary to pulmonary leukostasis
Although there is not extensive experience, imatinib has proved effective in treating pulmonary leukostasis. Hydroxyurea at doses of 3-4 g per day has also been used to treat pulmonary leukostasis but this patient developed disease progression on the drug and its beneficial effects in this regard would at best be temporary. Leukapheresis, if available, can be an effective means of lowering the leukocyte count temporarily and should be considered, particularly if there are neurologic signs or increasing pulmonary compromise. Cyclophosphamide can rapidly lower the leukocyte count but in this instance, imatinib, because of its target cell specificity, would be the drug of choice. Anemia is actually a protective mechanism when there is hyperleukocytosis and leukostasis, since it contributes to reducing whole blood viscosity. As blood viscosity is an exponential function of the hematocrit, red cell transfusion is contraindicated until the leukocytosis has been controlled.
A bone marrow aspirate and biopsy, and cytogenetics are obtained. The marrow is hypercellular with the expected left-shifted myeloid hyperplasia, including 20% blast cells. Micromegakaryocytes, pseudo-Gaucher cells are seen, stainable iron is present and reticulin is increased. However, there is a paucity of red cell precursors. The paucity of red cell precursors suggests the presence of red cell aplasia.Although the marrow histology is compatible with the accelerated phase of CML with respect to the myeloid criteria, the marked decrease in red cell precursors is not typical. Together with the very low reticulocyte count, the paucity of red cell precursors suggests that the patient is developing red cell aplasia. In this setting, red cell aplasia can be a harbinger of transformation. Overproduction of hematopoietic progenitor cells in a specific lineage-restricted pathway is not a cause of decreased production in other hematopoietic lineage-restricted pathways.
*** A 73-year-old man with type 2 diabetes mellitus is referred for evaluation of leukocytosis; a leukocyte count one year previously had been normal. Other than diabetes, which had required laser therapy for retinopathy and a complaint of fatigue, the patient had no other medical problems. Physical examination reveals a blood pressure of 140/95 mm Hg, a few retinal microaneurysms and a palpable spleen tip. Laboratory studies reveal a hemoglobin of 11.0 g %, an MCV of 98 fL, a leukocyte count of 76,400/μL with 12% myelocytes, 2% metamyelocytes, 18% bands, 56% neutrophils, 2% monocytes and 10% lymphocytes, and a platelet count of 366,000/μL. The bone marrow aspirate and biopsy are histologically compatible with chronic phase CML and a cytogenetics study reveals the Ph chromosome translocation as well as a del9. You start imatinib at a dose of 400 mg daily with 300 mg of allopurinol.
There is no evidence that the magnitude of the leukocyte count influences the effectiveness of imatinib and studies of the efficacy of higher doses of imatinib, as compared to the 400 mg dose, are still incomplete. The current standard of care in CML in the chronic phase is 400 mg of imatinib daily with 300 mg of allopurinol for hyperuricemia, (or to prevent it when the leukocyte count is elevated). Leukocyte count reduction with imatinib is sufficiently rapid that adjunctive therapy with hydroxyurea is unnecessary. The combination of hydroxyurea and imatinib could lead to significant neutropenia, resulting in the need to interrupt treatment or add growth factor support.
*** With treatment, the patient’s leukocyte count is reduced to normal and the platelet count is unchanged, but the hemoglobin level remains at 11.0 g % and the patient still complains of easy fatigability. A recent colonoscopy was normal and three stool specimens tested negative for occult blood. The reticulocyte count is 0.9%, the MCV 89, the serum creatinine is 1.0 mg %, a blood smear shows no abnormalities and a urinalysis is normal. Anemia? Why? Because the erythropoietin deficiency is secondary to underlying diabetes mellitus.
Anemia is a consequence of untreated CML and can also be a consequence of imatinib therapy but this patient’s leukocyte count has been brought under control and the platelet count remains normal, suggesting that neither CML nor imitanib is causing bone marrow suppression. Diabetes mellitus is a significant cause of anemia as a result of diminished erythropoietin production even in the absence of chemical evidence of renal impairment. Correction of anemia with erythropoietin therapy is indicated because the patient is symptomatic.
*** The patient’s hematologic response is excellent once his anemia has been corrected but now, a month later, he complains of periorbital edema and a severe headache. A CT scan shows cerebral edema.
Stop the imatinib! Imatinib is known to cause fluid retention manifest often as periorbital edema and weight gain and rarely as cerebral edema. Diuretics have not been effective in imatinib-induced cerebral edema. The imatinib must be stopped because imatinib-induced cerebral edema has been fatal. Retreatment with imatinib has also been associated with recurrence of the cerebral edema.
Lowering the dose of imatinib for minor degrees of toxicity (Grade 1-2) is unacceptable since this increases the possibility of losing the response and permitting the emergence of imatinib-resistant clones. For higher degrees of toxicity, stopping imatinib is mandatory until satisfactory resolution of the toxicity. Restarting at a lower dose may be necessary but not lower than 300 mg qd.
***A 46-year-old man with chronic phase CML complains of fatigue, anorexia and numbness and tingling of his fingers and toes. The patient has been taking interferon. The hematocrit is 33%, the MCV 110, the leukocyte count 25,000/μL. The platelet count is 100,000/μL and the reticulocyte count 0.05%. A red cell folate level is 200 nmol/mL. The serum vitamin B12 level is 545 pmol/mL. This is Pernicious anemia.
Myelodysplasia is usually a complication of CML treated with imatinib but neurologic symptoms are not a feature of myelodysplasia. In CML, there is an increase in the circulating level of transcobalamin I, a vitamin B12 carrier that does not deliver the vitamin to the tissues. Transcobalamin I can compete for available vitamin B12 with transcobalamin II, the physiologic carrier of vitamin B12, which is present in the circulation at a much lower concentration. Therefore, it is possible to have megaloblastic anemia from B12 deficiency without a decrease in the total serum B12 level. Folic acid is dependent on vitamin B12 for both its entry into the red cell and its storage there. When the level of available vitamin B12 that is carried on transcobalamin II is low, red cell folate will be low as well. Rarely, interferon therapy can cause a neuropathy but it would not cause macrocytosis.
*** A 40-year-old woman is referred for excessive bleeding following a tooth extraction. The patient claims general good health but does note that over the past year she has bruised easily and that her menses have become excessive, which she considers premenopausal. Physical examination is normal except for some scattered ecchymoses on the extremities. The blood counts reveal a hemoglobin of 11.0 g %, an MCV of 85 fL, a leukocyte count of 11,000/μL with 75% neutrophils, 5% monocytes, 18% lymphocytes, 2% basophils and a platelet count of 2,750,000/μL. Hmm? How about a ristocetin cofactor activity assay now? Now? What about NOW?
The patient’s hemorrhagic diathesis is manifested by surface hemorrhage and thus would not likely be caused by an abnormality of the extrinsic pathway. While acquired factor V deficiency has been documented in CML, this has usually been associated with very high leukocyte counts. The bleeding time can be normal or elevated with thrombocytosis and does not define the mechanism for hemorrhage. Essential thrombocytosis (ET) could well explain the patient’s clinical phenotype. However, the presence of basophilia would be distinctly unusual in ET and the JAK2 V617F mutation is present in only 45% of ET patients, making it a relatively insensitive test.
The patient’s hemorrhagic diathesis and the quality of the bleeding suggest a platelet or vascular defect, or von Willebrand’s disease. Given the recent onset of bruising and bleeding, the defect is likely to be an acquired one and given the marked thrombocytosis, it is most likely an acquired form of von Willebrand’s disease. With marked elevation of the platelet count, there is increased destruction of the largest and most hemostatically effective von Willebrand multimers, causing a type 2 form of von Willebrand’s disease. This is best identified by a ristocetin cofactor assay and values less than 30% are usually associated with an hemorrhagic diathesis. CML can present as isolated thrombocytosis but RT-PCR for BCR-ABL should not be used as the diagnostic test for CML because the test is not sensitive and false-positive results have been obtained in patients with essential thrombocytosis.
There is no correlation between the magnitude of the platelet count and major vessel thrombosis. In CML, in contrast to the other chronic myeloproliferative disorders, major vessel thrombosis has not been a significant complication associated with thrombocytosis. CML patients, however, are still at risk for the transient microvascular complication of thrombocytosis such as ocular migraine, erythromelalgia or transient ischemic attacks.
There is a direct correlation between the magnitude of the platelet count and the risk of bleeding because as the platelet count exceeds 1,000,000/μL, high molecular weight von Willebrand multimers are adsorbed and destroyed by the platelets, leading to an acquired form of von Willebrand’s disease. Since there is no correlation between the platelet count and thrombosis, reducing the platelet count, without reference to other possible risk factors for thrombosis, will not reduce the risk of arterial thrombosis.
*** A bone marrow aspirate and biopsy are performed on the patient for diagnostic purposes. The patient calls the next morning complaining of continual oozing of blood from the biopsy site, which has soaked her night gown. On physical examination, there is a large superficial ecchymosis at the biopsy site tracking down the right thigh and there is continuous oozing of blood from the biopsy puncture site. You give her epsilon aminocaproic acid.
The bleeding is not from lack of vitamin K-dependent clotting factors and an infusion of fresh frozen plasma would only expose the patient to allogeneic plasma. Activated factor VII is generally reserved for deep-seated hemorrhage secondary to a circulating inhibitor of coagulation or a deficiency of vitamin K-dependent factors such as occurs with a coumadin overdose and requires very prompt correction. Given the magnitude of the patient’s platelet count and the mechanism for the hemorrhagic diathesis, a platelet transfusion would either have no effect or would actually aggravate the problem.
Epsilon aminocaproic acid is an antifibrinolytic agent that stabilizes the platelet plug and is an effective hemostatic agent in von Willebrand’s disease when given either orally or intravenously. Alternatively, an infusion of DDAVP, which releases high molecular weight von Willebrand multimers from endothelial cells, could be tried. Both of these approaches would only be temporary. Plateletpheresis is an inefficient, temporary procedure, which often does not reduce the platelet count to the desired level and is subject to rapid rebound of the platelet count.
In the absence of other electrolyte abnormalities, the most likely cause for the hyperkalemia is potassium release during clotting of the blood sample as a result of the great mass of platelets in the sample. An ECG would not show tall, peaked T waves or a shortened QT interval but is probably unnecessary at this potassium concentration in an asymptomatic patient without evidence of cardiac, renal or endocrine disease. Since the most likely cause of the hyperkalemia is a test-tube artifact secondary to platelet potassium release upon blood coagulation, no therapy is necessary. A repeat serum potassium alone would only confirm the initial results.
A repeat serum potassium on a sample of anticoagulated blood will be normal, confirming the basis for the test-tube artifact.
*** A 44-year-old physician is noted to have a mild anemia and leukocytosis on a routine health maintenance evaluation. The patient has been asymptomatic and without prior medical problems. A year previously, his hemoglobin was 14.0 g %, the leukocyte count 3,700/μL with 74% neutrophils, 20.5% lymphocytes, 5% monocytes and 0.5% basophils, and the platelet count 194,000/μL. Physical examination is normal. Laboratory studies reveal a hemoglobin of 12.3 g %, an MCV of 88 fL, a leukocyte count of 21,140/μL with 49% neutrophils, 7% bands, 7% metamyelocytes, 16% myelocytes, 4% promyelocytes, 1% myeloblasts, 2% basophils, 8% monocytes and 6% lymphocytes, and a platelet count of 224,000/μL.
Although cytogenetic analysis is essential for monitoring therapy in CML, marrow cytogenetics is the important focus of this type of analysis and the lack of sufficient dividing cells may render peripheral blood cytogenetics less sensitive. Conventional cytogenetics also permits identification of other cytogenetic abnormalities that may be present such as trisomy 8 or isochromosome 17q. Peripheral blood BCR-ABL FISH, which assesses interphase nuclei, avoids the difficulties associated with a lack of dividing cells in the peripheral blood, but there is not the same record of experience with it as there is with conventional cytogenetics with respect to the effects of chemotherapy, and its degree of sensitivity is not that much greater. FISH analysis is useful when the Ph translocation is not detected by conventional cytogenetics and before therapy for identifying complex or variant translocations other than t(9:22)(q34;q11), Ph amplification and del9.
Real time quantitative RT-PCR for bcr-abl in the peripheral blood should be done before therapy to provide a baseline for monitoring residual disease during therapy. A bone marrow aspirate, biopsy and conventional cytogenetics should be performed for diagnostic purposes, and to provide a baseline for monitoring the effects of chemotherapy. Leukocyte alkaline phosphatase (LAP) is usually low in CML in contrast to polycythemia vera. However, the LAP score can be affected by disease status (increased in blast crisis), corticosteroid use, myelofibrosis and infections and may not be not depressed in CML presenting with isolated thrombocytosis.
A physician should discuss every aspect of the disease and treatment with the patients and their relatives. Disease stage and cytogenetic abnormalities are important prognostic factors, these all have to be discussed and explained. Early chronic phase without additional cytogenetic abnormalities offers the best prognosis, although there is still a small but finite risk of rapid conversion to blast crisis. Age, spleen size, platelet count and % peripheral blood myeloblasts have also been shown to predict the cytogenetic response to imatinib.
Allogeneic bone marrow transplantation is the only known curative therapy for CML. Let’s just say the patient has a healthy sister, who, if willing, should be tissue typed. A relative risk score for bone marrow transplantation has also been devised to aid in decision-making with respect to this procedure. Since the long-term effectiveness of imatinib has not yet been established and bone marrow transplantation has a finite immediate mortality risk and the possibility of substantial morbidity, patients preference become a vital piece of the therapeutic equation.
In 1984, Sokal and his colleagues analyzed the clinical features of 813 chronic phase CML patients to determine the prognostic significance of these features. Four clinical variables — age, spleen size, platelet count (>700,000/mL) and percent circulating blast cells — were found to segregate patients into low, intermediate and high risk groups with respect to their responsiveness to conventional chemotherapy therapy. However, the same risk factors appear to be useful for more modern forms of therapy such as interferon alpha and imatinib.
A common question is that weight matters. No. Clinical trial data has indicated that 400 mg per day is the effective imatinib starting dose for chronic phase CML. There is no need to adjust the starting dose of imatinib on a weight basis. The reduction in leukocyte count caused by imatinib primarily reflects suppression of the CML clone. When there is significant (grade 3-4) neutropenia, imatinib should be discontinued until there is neutrophil count recovery. Imatinib should never be reduced, as this increases the risk of selecting a resistant clone. With severe neutropenia, growth factor support can be employed to accelerate recovery. Imatinib can cause liver function abnormalities but it is not recommended that the drug be discontinued unless the increase in liver function tests is greater than 2.5-5 times the baseline normal value.
Myelofibrosis is a reactive, reversible process that is associated with a wide variety of benign and malignant disorders but whose presence makes it very difficult to define the underlying process causing it. Additionally, myelofibrosis is part of the terminal transformation process of many hematologic malignancies. Thus, the clinical picture described above is compatible with any of the listed diagnostic possibilities. However, the normal platelet count, the basophilia and the leukoerythroblastic reaction presence of a myeloproliferative process such as CML, which can present in blast crisis.
CML should always be a diagnostic consideration in patients such as this one because imatinib is an effective adjunctive therapy in combination with conventional chemotherapy when CML presents in this fashion, in contrast to the lack of success of conventional chemotherapy for the other illnesses listed above.
Typically, in CML, the peripheral blood contains the full spectrum of differentiating neutrophils from myeloblasts, promyelocytes, myelocytes, metamyelocytes and bands to mature neutrophils, with a greater proportion of myelocytes than metamyelocytes. An increase solely of neutrophils and band forms is more typical of the rare disorder, CNL, and an increase of monocytes of 1,000/μL or greater is usually considered a hallmark of CMMoL. IMF, of course, can present with a neutrophilic leukocytosis, anemia and splenomegaly and all of these diseases are most common in older men.
The balanced translocation between the long arms of chromosomes 9 and 22, which is both necessary and sufficient for the CML phenotype, is variable with respect to the location of the breakpoint in the bcr gene on chromosome 22. In most CML patients, the position of the breakpoint is such that the BCR-ABL fusion protein has a molecular mass of 210 kDa (p210). In some patients, however, the breakpoint gives rise to BCR-ABL fusion proteins with molecular masses of 230 kDa (p230) or 190 kDa.
These less common fusion proteins are associated with features that mimic CNL (p230) or CMMoL (p190). The patient described above fits the p185 phenotype. While rare, it is important to consider CML in the differential diagnosis of CNL, CMMol and IMF because of the availability of targeted therapy in the form of imatinib and other newer bcr-abl inhibitors such as dasatinib.
*** A 56-year-old man with chronic phase CML complains of the sudden onset of shortness of breath, pleuritic chest pain and a low grade fever. The patient had been intolerant of imatinib as the result of intractable diarrhea and was started on interferon alpha. However, because of a leukocyte count of 85,000/μL and night sweats, he was also started on hydroxyurea and allopurinol. Three days after starting the hydroxyurea and interferon, he presents with complaints of the sudden onset of left upper quadrant pain, pleuritic chest pain, dyspnea and fever. The physical examination reveals a dyspneic, diaphoretic man with a temperature of 100.4°F, respirations of 20/minute, a blood pressure of 140/90 and a pulse rate of 120 beats per minute. There is tenderness over the left trapezius muscle, and respiratory splinting with dullness and diminished breath sounds over the left lower lobe. There was a sinus tachycardia but no murmurs or gallops were heard. The spleen was easily felt at 10 cm below the left costal margin with marked tenderness laterally. A rub was heard over the 11th rib laterally. The liver was not enlarged and bowel sounds were present. The extremities showed no edema or cords. The hemoglobin is 12 g %, the leukocyte count 65,000/μL and the platelet count 465,000/μL. A urinalysis is normal. A chest x-ray reveals elevation of the left diaphragm and a small left pleural effusion. Abdominal x-rays reveal no free air. An ECG is normal and the arterial oxygen tension is 95 mm Hg.
Pulmonary embolism with infarction is always a diagnostic consideration with this type of clinical presentation but usually pulmonary infarction occurs in the setting of cardiac failure, which is not the situation here. This is a classical presentation for splenic infarction, which occurs when the spleen outgrows its blood supply. In this instance, the patient’s disease had not yet been brought under control with hydroxyurea and interferon. The possibility of pneumonia must always be considered with this clinical presentation but the absence of a pulmonary infiltrate makes this diagnosis unlikely. Gastric perforation should be included in the differential diagnosis, particularly with the signs of diaphragmatic irritation but the absence of free air on x-ray and the presence of bowel sounds make the diagnosis unlikely. A uric acid stone could present acutely with abdominal pain but the presence of pulmonary symptoms and signs, a normal urinalysis and the administration of allopurinol make this diagnosis unlikely.
This patient has Splenic infarction.
Only ionizing radiation is a risk factor for the development of CML.
- Drug-induced toxic epidermal necrolysis (TEN) or the Stevens-Johnson syndrome, which is a less severe form of TEN, are immunologically-mediated drug reactions and not dose-related.They thus require immediate and permanent withdrawal of the offending drug and institution of steroid therapy as well as intensive management of the skin desquamation.
- Low dose interferon alpha therapy (3 MIU three times a week) is equivalent to and better tolerated than 5 MIU given daily. Unlike imatinib, the effects of interferon on the bone marrow persist after it is discontinued and it can be given safely during pregnancy. When discontinued for more than 90 days, interferon therapy does not interfere with bone marrow transplantation.
- No organ system is spared interferon toxicity but many of its side effects develop insidiously and are easy to overlook initially. This is particularly true for hypothyroidism and renal failure. Other side effects such as atrial fibrillation can occur suddenly.
***A 19-year-old man complains of a persistent, painful penile erection for over 19 hours. The patient has been in otherwise good health and denies the use of any performance-enhancing drugs. A complete blood count reveals an hematocrit of 39%, a leukocyte count of 380,000/μL with a differential count of 47% neutrophils, 11% bands, 5% metamyelocytes, 19% myelocytes, 5% promyelocytes, 2% blast cells, 5% monocytes and 6% lymphocytes and a platelet count of 147,000/μL.
Priapism is a rare complication of leukostasis in CML, which, if not reversed promptly, can result in erectile dysfunction. The most successful therapeutic approach to the resolution of priapism involves both systemic and local therapy. Thus, in the clinical situation described above, local aspiration and irrigation, followed by phenylephrine injection, should be initiated immediately while leukapheresis is being arranged. Hydroxyurea should be administered concurrently to insure that there is no rebound leukocytosis.
Thanks for reading,