One of the headlines this week was that NICE has approved thalidomide for the treatment of myeloma. Since thalidomide was such a bete noir of the investigative press all those years ago, it was worth a mention on the BBC News. Today there is a good article on thalidomide and its derivatives, lenalidomide and pomalidomide (collectively known as IMiDs), in Cancer Treatment Reviews from the German group that includes Stephan Stilgenbauer and Harmut Dohner, of whom I am such a big fan. The three diseases that they are valuable in, Myeloma, MDS and CLL, are the three diseases that I have spent most of my life studying.
The anti-inflammatory role of thalidomide was discovered by chance after giving it as sedative to a leprosy patient suffering from erythema nodosum leprosum. The underlying molecular mechanism was later shown to involve inhibition of the B-cell receptor, which is necessary for antigen-induced activation of the immune system.
In contrast, Rho GTPase-dependent intracellular signaling seems to be activated by pomalidomide, a third-generation IMiD, resulting in reorganization of the cytoskeleton and enhanced cellular migration.
In multiple myeloma there seems to be a direct cytotoxic effect of immune modulatory drugs on tumor cells that are resistant to conventional therapy. It is mediated via inhibition of NFκB and induction of caspase 8 activity. In contrast, in CLL cells an induction of NFκB and NFAT-signaling has been reported for lenalidomide together with activation of PI3K-activity, resulting in enhanced transcription and mRNA stabilization of the CD154 receptor that itself also mediates activation of the canonical and non-canonical NFκB signaling pathway. Completely opposite effects in CLL and myeloma! In CLL, the induction of NFκB leads to increased expression of pro-apoptotic molecules like BID and DR5 which sensitize CLL cells to TRAIL-induced apoptosis, but also seems to reverse the humoral tolerance against tumor specific antigens like ROR1, suggesting a dual mode of action.
In the minimally deleted region on chromosome 5q in MDS, two phosphatases (CDC25C and PP2A) have been identified that control the transition of G2 to M in the cell cycle. These dual-specificity serine threonine phosphatases are deleted in the majority of del 5q MDS patients and are also inhibited in their enzymatic activity by lenalidomide. The subsequent inhibition of cell proliferation could be the reason why del 5q MDS patients are especially sensitive to treatment with the drug. The expression of the tumor suppressor gene SPARC that is also localized in the critical region on 5q. Incubation of del 5q cells, which have low levels of SPARC, with lenalidomide upregulated the gene back to normal levels. The proposed function of SPARC is as a cell cycle inhibitor and counteradhesive factor.
Another cell cycle inhibitor that is upregulated by lenalidomide in Burkitt-lymphoma and MM-derived cell lines is the CDKN1A/p21/WAF gene. I am afraid that the mechanism for this happening, which involves epigenetics is too complicated to explain to you. I must ask my daughter what it means.
An elevation of TNFa production is implicated in the pathogenesis of a number of hematopoietic malignancies including CLL. TNFa levels are reduced after treatment
with lenalidomide as early as 7 days after the beginning of therapy. This reduction correlates with a decrease in the white blood cell count.
Thalidomide and its derivatives may influence cytokine patterns differently, allowing them to be grouped into two different classes depending on whether they increase levels of IL-2, IL-10, IFN gamma, and the activity of phosphodiesterase 4 and decrease of TNFa, IL-1ß and IL-6. IL-6 is known to inhibit the apoptosis of malignant myeloma cells and help in their proliferation. In-vitro, the production of IL-6 is directly downregulated by lenalidomide, which results in the reduction of osteolysis and changes in the tumor microenvironment. In MM patients however, there is an increase in levels of IL-6 upon treatment with lenalidomide, and IL-6 (together with IL-10, IL-2 and TNFaR1) is also increased on day 7 after treatment with lenalidomide in patients with CLL , suggesting a transient immune activation in these patients.
Changes in cytokine patterns subsequently lead to changes in the composition of
the lymphocyte compartment, which had already been observed in patients suffering from leprosy, where the T-helper cell compartment was increased after treatment with thalidomide. Thalidomide has been shown to act as costimulator to CD3 engagement, activating CD8+ T-cells.
However, the activation of T-cells is dependent on the IMiD used, with one group of second-generation drugs lacking this activating capacity. In-vitro, immune modulatory drugs act on T-cells via the B7-CD28 co-stimulatory pathway. Blockade of this interaction can be overcome by incubation with IMiDs, but not via upregulation of B7 or CD28 on T-cells or antigen presenting cells (APC), rather, they induce phosphorylation of CD28, resulting in activation of downstream targets like PI3K and NFκB and subsequent activation of the T-cell. This T-cell stimulation results in increased secretion of IFN gamma and IL-2, the Th1-type cytokine response, and can stimulate clonal T-cell proliferation and NK cell activity.
In 50% of patients with MDS but only in 5% of age-matched controls, a clonal expansion of effector T-cells with natural killer (NK) surface markers (e.g. NKG2D) is observed by analyzing T-cell receptor clonality at the CDR-3. In patients with MDS, CLL or MM, treatment with lenalidomide results in an antigen-dependent increase in NK cell populations. In addition, in vitro experiments have also shown an increased killing capacity of NK-cells towards MM cells after treatment with IMiDs. Both thalidomide and lenalidomide also seems to enhance the immune response in CLL patients by downregulating the number of suppressive regulatory T-cells (CD4+ CD25high FOXP3+). Similarly, pro-inflammatory IL-17 positive regulatory T-cells are upregulated in CLL patients at a later stage after treatment with lenalidomide.
In CLL the formation of the immunological synapse between the malignant B-cells and the T-cells is impaired, possibly due to aberrant actin polymerization and subsequent
reduction of polarity of the T-cells. This lack of polarization is also induced in T-cells from healthy donors by CLL cells. However, upon treatment with lenalidomide, the in vitro capacity of T-cells to form an immunological synapse can be recovered.
One of the key elements of the pathology of hematological diseases is the interaction of the malignant cells with the bone marrow. The adhesion of tumor cells to bone marrow stromal cells often induces the secretion of pro-angiogenic and other cytokines, which are essential for the pathogenesis of the disease.
Bone marrow stromal cells (BMSC) are required for normal hematopoiesis, but also give pro-survival and proliferative support to MM cells. The close interaction between MM cells and BMSCs is partially based on the induction of secretion of IL-6 and VEGF from BMSCs by VEGF secreted from MM cells. This interaction can be inhibited by neutralizing antibodies directed against IL-6 or VEGF, but inhibition is more practically induced clinically by treatment with thalidomide.
In contrast, in 5q- MDS it seems that IMiDs do not directly impact on the microenvironment, rather that eradication of the MDS clone leads to restoration of normal stromal function.
Malignant cells from the peripheral blood and lymph node aspirates of CLL patients undergo rapid apoptosis when taken out of the patient and cultured in vitro, a process which can be rescued by co-culture with stromal cells in vitro. In addition, a reduction of the immunocompetence of non-malignant cells can be observed upon co-culture with CLL cells. These interdependencies of CLL cells with their stromal microenvironment may be one of the reasons why IMiDs are effective in the treatment of this disease, blocking the pro-survival support and possibly reinstating the suppressed immunocompetence of the CLL microenvironment. It remains to be seen whether the observed tumor flare reactions that occur in the lymph nodes of CLL patients upon treatment with high doses of lenalidomide is due to the re-establishment of the immune competence of the microenvironment in these organs.
In MM, there is increased bone marrow microvascular density and VEGF levels. Both thalidomide and lenalidomide inhibit angiogenesis.
As patients with del5q MDS respond best to lenalidomide compared to patients without the deletion, levels of VEGF and its receptor KDR were analyzed in MDS patients responding to lenalidomide treatment. In line with VEGF signaling being a major target for IMiD function, both VEGF and KDR were found to be significantly
downregulated in MDS patients responding to treatment and this downregulation proved to be a prognostic factor for induction of remission.
Co-culture of CLL cells on stromal cells induces a pro-angiogenic phenotype in CLL cells including enhanced expression of VEGF, a process that is reminiscent of the interaction of MM cells with their BM microenvironment. The exact molecular mechanism remains unclear.
Early trials of third generation IMiDs like pomalidomide are promising. In a phase I clinical trial of pomalidomide treatment in relapsed MM patients, only 3/24 patients developed thrombosis, the treatment-related non-hematologic dose-limiting toxicity. Thalidomide-resistant patients were allowed, but still the ORR was 53% with four CRs.
An even more promising therapeutic strategy is the combination of IMiDs with compounds targeting either different proteins within the same pathways like dexamethasone, or the combination with therapeutics that target different pathways and therefore act synergistically.