Bruce A. Chabner
An obvious feature of malignant cells is their failure to differentiate, and to acquire the histologic, biochemical, and functional features of the mature cells of the tissue from which they arise. Thus, leukemia cells resemble in appearance, surface markers, and molecular profile the more primitive normal progenitors of the myeloid or lymphoid series. Indeed it has been possible to isolate a small fraction of continuously self-renewing cells (called tumor stem cells) from frankly malignant tissues. These tumor cells are able to reproduce multiple differentiated cell lineages when appropriately stimulated by “differentiating” agents such as 5-azacytidine (see Chapter 1) or retinoids. The progression from mature cell phenotype to an undifferentiated malignancy is recapitulated in serial observations of chronic myelogenous leukemia and in experimental models of malignant transformation.
Research efforts have defined pathways responsible for a block in differentiation in malignant cells and have suggested strategies for pharmacologic intervention. Vitamin A and related retinoids were the first compounds to show differentiating effects in cell culture, and all-trans retinoic acid (ATRA) (Figure 9-1) was subsequently found to be highly effective in inducing remission in promyelocytic leukemia, a disease characterized by a translocation involving the retinoic acid receptor, RAR-alpha (1). Subsequently, other pathways have been exploited as targets for development of differentiating agents, including histone deacetylase (HDAC), DNA cytosine methyltransferase (CMT), and vitamin D signaling pathways. CMT is targeted by both 5-azacytidine and decitabine, as discussed in the section on antimetabolites, while HDAC inhibitors and vitamin A analogs are useful agents in peripheral T-cell lymphoma and acute promyelocytic leukemia (APL), respectively. Here we will consider four clinically useful agents that promote differentiation: ATRA and arsenic trioxide (ATO), both of which are effective in APL, and vorinostat and rhombedepsin, which are approved for treatment of cutaneous T-cell lymphoma (CTCL).
FIGURE 9-1 Differentiating agents useful in APL therapy. (A) All-trans retinoic acid (ATRA) and (B) arsenic trioxide.
ATRA
Retinoids (vitamin A and its derivatives) induce differentiation of malignant cells in cell culture systems. Early work showed that retinoids caused promyelocytic leukemia (APL) (HL-60) cells to undergo maturation into granulocytes at drug concentrations easily achievable in humans. In normal cells, RAR-alpha forms homodimers as well as heterodimers with retinoid X receptor, and the dimer in turn complexes with the PML transcription factor. The RAR-alpha homodimer in complex with PML in turn binds ATRA, leading to chromatin modification and transcription of genes that induce differentiation (2). In APL cells, the normal pathway for vitamin A action is disrupted by a translocation fusing portions of the RAR-alpha gene on chromosome 15 and the PML gene on chromosome 17. The fusion protein acts as a repressor of differentiation and, through the FOS gene, promotes proliferation. The RAR-alpha/PML fusion protein has low affinity for ATRA and requires pharmacologic concentrations of retinoid to activate differentiation in APL cells. High concentrations of ATRA lead to multiple effects on leukemic cell differentiation: degradation of the fusion protein, chromatin modification, and upregulation of transcription factors (CEBP beta, OCT-1, and most importantly, PU1) required for myeloid differentiation (2). Resistance to ATRA arises through mutation of the fusion protein to decrease its binding affinity, and possibly by induction of the ATRA metabolizing p450 isoenzyme (CYP26A1) in the liver or in APL cells after prolonged exposure to the drug (3).
CLINICAL PHARMACOLOGY
ATRA induces complete remission in more than 90% of patients with APL, alone or in combination with an anthracycline (1). The initial clinical experience with ATRA revealed clear evidence of leukemic cell maturation in bone marrow and peripheral blood. Remission duration for single agent ATRA tends to be brief, and the recurrent tumor is usually refractory to a second cycle of treatment due to further mutation in the RAR-alpha gene. However, when ATRA is used in combination with anthracyclines as primary therapy, the regimen is curative for 70% or more of patients with this disease. It is effective in all patients with the PML/RAR-alpha translocation, including those with a masked translocation, only detectable by PCR. The “long” form of the translocation is more sensitive to APL and has a greater disease-free survival after remission than does the “short” form (4).
ATRA is given in daily intravenous doses of 45 mg/m2/day until remission is achieved. It is now used as the initial therapy in APL in combination with an anthracycline. ATRA has the additional advantage of quickly aborting the life-threatening coagulopathy associated with this disease. In patients with significant comorbidities or in elderly patients, it efficiently induces remission when used with arsenic trioxide, without a cytotoxic agent. It may also be effective in maintenance therapy, but its continuous use leads to cutaneous toxicity and elevated plasma triglycerides. Therefore, intermittent schedules of maintenance ATRA, given with methotrexate and 6-mercaptopurine, may be better tolerated and equally efficacious. ATRA concentrations in plasma reach 400 ng/ml, and ATRA is rapidly eliminated by p450-mediated metabolism. The half-life in plasma is less than 1 h, and the rate of clearance increases markedly with repeated doses.
TOXICITY
ATRA’s primary side effects are cutaneous and pulmonary. Retinoids cause a redness, dryness, and sensitivity of the skin and cracking of the lips (cheilosis). A more important, and at times lethal, toxicity is its tendency to cause pulmonary dysfunction. Through its differentiating effects on APL cells, it causes an accumulation of leukemic myeloid precursors in pulmonary vessels, leading to hypoxia, pleural and pericardial effusions, peripheral edema, fever, and, in extreme cases, respiratory failure and death. Concurrent glucocorticoids and cytotoxic chemotherapy (anthracyclines), given with ATRA, control the increase in leukemic granulocytes in the peripheral blood, decrease their adhesion to endothelium, and greatly lessen the risk of the “retinoic acid syndrome.” Prophylactic dexamethasone should be given to patients with a presenting white blood cell count of 5 × 109/l or higher (5).
Other ATRA toxicities include pseudotumor cerebri, with headaches, changes in mental status, and papilledema; liver function test abnormalities; bone tenderness; hypercalcemia and renal failure; myocarditis; and elevated plasma triglycerides, all of which reverse with cessation of therapy.
ARSENIC TRIOXIDE (ATO)
A second unique therapy, ATO (Figure 9-1) effectively treats APL, with differentiating action that mimics ATRA. ATO induces a high rate of complete response in patients refractory to ATRA and chemotherapy, and is effective as a component of consolidation therapy (6) and in elderly patients during remission induction (7). It promotes differentiation and apoptosis of cultured APL cells. It causes the degradation of RAR-alpha/PML through production of reactive oxygen species, which induce intramolecular disulfide linkages in the PML portion of the fusion protein. ATO binds directly to the disulfide-linked PML, and the complex transfers to the nuclear matrix, where it undergoes sumoylation, ubiquitination, and degradation (8). ATO also promotes degradation of NF-κB, an antiapoptotic transcription factor, and suppresses key transcription factors in the hedgehog signaling pathway (9). The contribution of each of these properties to the antileukemic action of APL is not clear.
CLINICAL PHARMACOLOGY
The standard regimen for ATO administration as a single agent is a 1- to 2-h infusion of 0.15 mg/kg/day for up to 60 days or until achievement of a complete response. Further treatment is given after a 3-week break. Alternative schedules of a 5-day loading dose, followed by twice weekly drug infusions until remission, are being explored. Its use in combination with ATRA and cytotoxic chemotherapy for primary therapy of APL is evolving. Early trials indicate that it is highly effective in inducing remission when used with ATRA, both in low-risk and high-risk patients (WBC count > 5 × 109/l), with minimal toxicity (7).
Peak concentrations of total arsenic achieved during the 2-h infusion reach 5 μM. The parent compound is eliminated through interaction with sulfhydrils and through enzymatic methylation. The concentration of parent drug in plasma, the active principle, is probably lower than 1 μM (10).
TOXICITY
ATO causes a long list of side effects, the most important of which is a leukemic cell maturation syndrome similar to that caused by ATRA, with pulmonary distress, pleural and pericardial effusions, and alteration in mental status. This syndrome is effectively prevented and treated with dexamethasone, 10 mg, which should be administered concurrently with ATO in patients with white blood cell counts greater than 5 × 106/ml at presentation. ATO may cause hyperglycemia, hepatic enzyme abnormalities, and rarely acute hepatic failure. Myositis manifested as muscle tenderness and muscle swelling, accompanied at times by fever, has also been reported. ATO inhibits ion channels in the cardiac conduction system, causing a prolongation of the QT interval and predisposing to atrial and ventricular arrhythmias (torsade de pointes). During ATO therapy, a weekly EKG should be monitored for signs of QT prolongation greater than 500 ms and for arrhythmias. Serum K+ and Mg2+ should be monitored weekly and replenished as necessary to maintain concentrations above 4 meq/l (K+) and 2 meq/l (Mg+), respectively. An absolute QT interval of > 500 ms should lead to drug discontinuation and immediate repletion of electrolytes.*
HISTONE DEACETYLASE (HDAC) INHIBITORS
VORINOSTAT
The most recent additions to the list of differentiating agents approved for clinical use are two HDAC inhibitors, vorinostat (12) and romidepsin (13). HDACs are a large family of enzymes that remove acetyl groups from amino groups of the lysines found in chromatin and thereby produce compaction of chromatin, blocking gene transcription and differentiation. Inhibitors of HDACs reverse this process, promoting the transcription of DNA, blocking cell cycle progression, and leading to terminal differentiation and apoptosis. These inhibitors also alter the stability of a broad class of cell cycle checkpoint proteins and DNA repair proteins by blocking their deacetylation. HDAC inhibitors are indicated for treatment of cutaneous and peripheral T-cell lymphomas.
Vorinostat was approved based on its ability to cause partial or complete responses in 30% of patients with CTCL after failure of at least two prior regimens (11). Responses were achieved after a median of 55 days of treatment on a schedule of 400 mg per day, and lasted a median of 5.5 months. Vorinostat has a plasma half-life of 1.5–2 h. It is eliminated by glucuronidation and by hydrolysis and beta-oxidation. Asian patients with the UDP-glucuronyltransferase 2B17 genotype have delayed drug clearance and a higher rate of toxicity (14). Its primary toxicities are mild to moderate fatigue, anorexia, nausea, diarrhea, thrombocytopenia, and anemia. Serious or dose delaying side effects are uncommon, the most notable being thrombocytopenia in 6%. While the HDAC inhibitors as a class cause lengthening of the QT interval, there is no consistent evidence for cardiotoxicity or arrhythmias related to vorinostat.
ROMIDEPSIN
Romidepsin (depsipeptide), a complex natural product composed of unusual amino acids in a cyclic peptide linkage, also inhibits HDACs and is approved for treatment of CTCLs. It is similar to if not more potent and more clinically active than vorinostat, but it has consistent effects on the electrocardiogram (flattening of T-waves, modest prolongation of the QT interval). While in early trials two patients died, possibly due to drug-induced arrhythmias, further trials found that the drug is safe in routine clinical use at a dose of 400 mg per day. Monitoring of serum K+ and Mg+, repletion of electrolytes, and monitoring of the QT interval prior to drug administration are advised. Mild myelosuppression may also occur during prolonged use. Other toxicities are nausea, anorexia, and diarrhea. The drug has a plasma half-life of 3 h and is eliminated by CYP3A4 metabolism.
Both vorinostat and romidepsin may inhibit warfarin clearance and prolong the prothrombin time.
REFERENCES
1. Sanz MA, Tallman MS, Lo-Coco F. Practice points, consensus, and controversial issues in the management of patients with newly diagnosed acute promyelocytic leukemia. The Oncologist. 2005; 10: 806–814.
2. Mueller B, Pabst T, Fos J, et al. ATRA resolves the differentiation block in t(15; 17) acute myeloid leukemia by restoring PU l expression. Blood. 2006; 107: 3330–3338.
3. Idres N, Marill J, Chabot G. Regulation of CYP26A1 expression by selective RAR and RXR agonists in human NB4 promyelocytic leukemia cells. Biochem Pharmacol. 2005; 10: 1595–1601.
4. Tussie-Luna MI, Rozo L, Roy AL. Pro-proliferative function of the long isoform of PML-RARα involved in acute promyelocytic leukemia. Oncogene. 2006; 25: 3375–3386.
5. Wiley JS, Firkin FC. Reduction of pulmonary toxicity by prednisolone prophylaxis during all-trans-retinoic acid treatment of acute promyelocytic leukemia. Australian Leukaemia Study Group. Leukemia. 1995; 9: 774–778.
6. Powell BL, Moser B, Stock W, et al. Arsenic trioxide improves event-free and overall survival for adults with acute promyelocytic leukemia: North American Leukemia Intergroup Study C9710. Blood. 2010; 116: 3751–3757.
7. Hu J, Lio Y-F, Wu C-F, et al. Long-term efficacy and safety of alltrans retinoic acid/arsenic trioxide-based therapy in newly diagnosed acute promyelocytic leukemia. Proc Nat Acad Sci USA. 2009; 196: 3342–3347.
8. Jeanne M, Lallemand-Breitenback V, Ferhi O, et al. PML/RARA oxidation and arsenic binding initiate the antileukemia response of As2O3. Cancer Cell. 2010; 18: 88–98.
9. Platanias LC. Biological responses to arsenic compounds. J Biol Chem. 2009; 284: 18583–18587.
10. Fukai Y, Hirata M, Ueno M. Clinical pharmacokinetic study of arsenic trioxide in an acute promyelocytic leukemia patient: speciation of arsenic metabolites in serum and urine. Bio Farm Bull. 2006; 29: 1022–1027.
11. Mari F, Bertol E, Fineschi V, Karch S. Channelling the emperor: what really killed Napoleon? J R Soc Med. 2004; 97: 397–399.
12. Lane AA, Chabner BA. Histone deacetylase inhibitors in cancer therapy. J Clin Oncol. 2009; 27: 5459–5468.
13. Bertino EM, Otterson GA. Romidepsin: a novel histone deacetylase inhibitor for cancer. Expert Opin Investig Drugs. 2011; 20: 1151–1158.
14. Wong NS, Seah EZ, Wang ILZ, et al. Impact of UDP-glucuronyltrasferase B17 genotype on vorinostate metabolism and clinical outcomes in Asian women with breast cancer. Pharmacogenet Genomics. 2011; 11: 760–768.
*On a historical note, Napoleon appears to have been the victim of arsenic cardiac toxicity. An analysis of Napoleon’s hair has demonstrated high levels of arsenic, indicating chronic arsenic poisoning; it is believed his acute fatal episode was a ventricular arrhythmia (torsades de pointes) induced by hypokalemia that resulted from treatment with emetics and cathartics (11) given for his chronic gastrointestinal symptoms. At autopsy he was discovered to have a gastric carcinoma.