Discuss the efficacy and safety of haematopoietic stem cell (HSC) transplantation for multiple myeloma, including the pros and cons of autologous and allogeneic HSC
The difference between embryonic and tissue stem cells
Stem cells (SCs) were first suggested by McCulloch and Till (1960) before being further described and classified by Lajtha (1979). Here they were described as being able to self-renew and proliferate unchanged, within the life cycle of an organism. They were classified alongside cells in transient development and fully differentiated cells which were in a non-growing phase. The stem cells described in that paper are what we would consider today adult stem cells. These are multipotent, meaning they can differentiate in several different tissues, but not all tissue types (Watt and Driskell, 2010).
We now understand there are several types of SC which can differentiate in to greater or fewer tissue types. Totipotent cells can differentiation into all known tissue types, including those of the placenta and other extraembryonic tissues. Pluripotent cells can different into all known tissue types present in an adult organism. Multipotent cells are those previously described by Lajtha (1979) and differentiate into all cell types of a specific tissue. Unipotent cells are those that only differentiate into one cell type, usually to maintain a particular cell line.
Embryonic stem cells (ESCs) can be isolated from human blastocysts (Thomson et al., 1998) and can be cultured while still undifferentiated. ESCs are pluripotent and can differentiate into all cell types found in an individual adult (Vazin and Freed, 2010). It has been suggested that pluripotent SCs might be possible to source from human germ cells, in which case, some of the ethical controversies over the use of embryo-derived cells could be avoided (Kerr et al. 2006).
Adult tissues can also provide a source of SCs. The role of these cells in human development is primarily homeostatic, to maintain and renew tissue cells as part of normal cellular processes and apoptosis. Adult stem cells (ASCs) are often referred to as tissue SCs given, they most often, reside within body tissues (Robey, 2000). ASCs are found within the specific tissues that they maintain and generally, display multipotent differentiation. However, more recent studies have elucidated the process of reprogramming such tissue stem cells to become pluripotent or induced Pluripotent Stem Cells (iPS). At first this required the use of viral vectors and several transcription factors (Takahashi et al., 2007). Following this, the process of producing murine iPSC without the need for viral vectors has been demonstrated (Okita et al., 2008). Figure 1 shows a diagram and comparison of different stem cell types, their sources and potential for differentiation.
Haematopoietic stem cells (HSCs) are amongst the best characterised stem cells and are the progenitors to all mammalian blood cell types (Wilson and Trumpp, 2006). HSCs possess not only multipotency but also the ability to self-renew, to generate more HSCs without differentiation. As mature blood cells are constantly recycled and the role of HSCs in renewal and differentiation of these cells and tissues is critical (Watt and Driskell, 2010). HSC-derived clinical treatments are critical aspects of cancer care and have been used in both genetic (Shizuruet et al., 2005 and blood (Steward and Jarisch, 2005) disease treatments (Hawley, Ramezani and Hawley, 2006).
As HSCs can be found in both bone marrow and peripheral blood, they are readily accessible. Stem Cells can be treated with granulocyte colony stimulating factor (G-CSF) to mobilise them to enter the peripheral blood supply. Once collected, they can be genetically altered ex-vivo possibly allowing for the development of curative treatments for genetically originated blood disorders, for example the treatment of haemophilia which has been demonstrated in mice (Moayeri, Hawley and Hawley, 2005). Other sources of HSCs are bone marrow and cord blood and have differing properties regarding their usefulness in therapeutic applications (Haspel and Miller, 2008). Bone marrow residing HSCs have been shown to result in great quality of life measurements and a reduction in the chance of developing graft-versus-host disease (GVHD) when compared to the sue of peripherally coursed SCs however, the use of peripherally-sources SCs has been liked to improved clinical outcomes as a result of improves rates of cell proliferation and a reduced chance of graft failure, which in turn reduces the chance of disease relapse (Amouzegar, Dey and Spitzer, 2019) Figure 2 shows the process and outcomes of differentiation in HSCs.
The use of HSC therapies is associated with a number of general complications that must be appropriately planned for, considered and balanced. For example: Cytokine release syndrome (CRS), veno-occlusive disease, drug-induced microangiopathy, renal impairment, GVHD, pulmonary complications and hepatic sinusoidal obstruction syndrome (Uptodate.com, n.d.).
The efficacy and safety of haematopoietic stem cell (HSC) transplantation for multiple myeloma
Multiple myeloma (MM) is a malignant disease caused by an increase plasma cells in bone marrow. monoclonal immunoglobulins are often released into blood in the serum and may even appear in the urine (Röllig, Knop and Bornhäuser, 2015). MM represents 1% of all cancers globally and up to 15% of all haematological malignancies (Rajkumar et al., 2014). Over 30k new annual cases are recorded in the USA alone, with more than 12k MM patients dying as a result (Siegel, Miller and Jemal, 2018). It is the 3rd most common blood disease with a median onset age of 65 (Kyle et al., 2003). MM occurs more often males compared to females and presents in Caucasians with the half the frequency as in Afro-Caribbean populations (Landgren and Weiss, 2009). In addition to bone disease, which represents the primary cause of death (Roodman, 2008), MM is associated with higher incidence of infection, anaemia, elevated blood calcium levels and kidney dysfunction.
Most MM patients have experienced disease progression from a precursor state, referred to as “monoclonal gammopathy of undetermined significance” or, MGUS (Landgren et al., 2009). MGUS is asymptomatic and often goes undiagnosed for over a decade (Therneau et al., 2012). It is thought to be identified in up to 5% of people over 50 (Dispenzieri et al., 2010) and develops into MM in about 1% of cases annually (Kyle et al., 2018). MM can also develop from a more clinically advanced state than MGUS termed, “smouldering multiple myeloma” or, SMM (Kyle et al., 2007). SMM to MM has an annual disease progression of around 10% (Rajkumar, Landgren and Mateos, 2015). Figure 3 shows the diagnosis and progression of MM from MGUS and SMM precursor states.
Traditional MM treatment, especially for those with systemic MM and end-organ damage is whole-body chemotherapy. This halts disease development and can alleviate some MM-induced symptoms. Partial or complete remission following this chemotherapy programme generally leads to positive outcomes in the long term. As chemotherapy is considered a successful treatment option, treatment development aims to increase response rates and tolerability. (Palumbo et al., 2012). Quality of life is also a significant consideration, particularly as MM is not described as being curable (Röllig, Knop and Bornhäuser, 2015).
HSCs in combination with chemotherapy, have used in the treatment of MM for many years (Jagannath and Barlogie, 1992). Patients eligible for Haematopoietic Cell Transplantation (HCT) first undergo an induction therapy, which can last upwards of 4 months. This aims to lower the tumour burden, decrease MM symptom severity and prevent end-organ damage. Induction periods for longer periods of time, can be to the detriment of HSC collection (Kumar et al., 2009).
Autologous HSC Treatment (AutoHCT)
Peripheral blood progenitor cells (PBPCs) are collected and reinfused into the patient soon after the end of a chemotherapy cycle. As this results in a reduction in all types of blood cells, Erythrocyte and platelet transfusions can be provided if required. In addition, haematopoietic colony-stimulating factors (G-CSFs) are provided to improve engraftment (Schmitz et al., 1996). The transfusions described are not significant and some AutoHCT have been carried out without this supportive therapy (Joseph et al., 2018).
In individuals younger than 70, high-dose chemotherapy in advance of AutoHCT is the standard of care. While this is not curative, it is shown to improve both event-free (EFS) and overall survival (OS) (Palumbo et al., 2014). Although, some studies did not support this view, they did not directly compare HCT against chemotherapy alone (Dhakal et al., 2018).
As this is not curative, many patients will experience relapse of MM and those who do so within a year of AutoHCT have been shown to have a reduced OS compared to those who relapse after the first year of treatment (Kumar et al., 2008). The only course of action for relapsed MM patients is a second course of AutoHCT treatment, non-myeloablative Allogenic HCT (AlloHCT) or treatment with conventional chemotherapeutic agents (Rajkumar, Kyle and Connor, 2019). AutoHCT efficacy has yet to be demonstrated in renally impaired patients and this is clearly an area that requires further study and caution.
Engraftment is a fundamental aim of HSCs transplantation. It is needed for the treatment to be both long-term and to lead to the restoration of haematopoiesis in the patient. It is therefore the key measure used to assess OS improvements (Hutt, 2017). Delayed engraftment (DE) is an issue also seen in Allogenic HCT (AlloHCT) (Tricot et al., 1998), although the presence of DE in AutoHCT treatment is by no means conclusive (Stewart et al., 2001).
AutoHCT products are extremely difficult to purify and can often be contaminated with disease cells and MGUS or SMM precursors which, if infused can lead to disease re-emergence (Zhou et al., 2003) and has been confirmed by investigations in an MM murine model (Pilarski et al., 2000).
Patients are as elevated risk of infections and AutoHCT grafting can lead to depressed immune response and cytopenia. It is thought that up to 40% of AutoHCT patients may develop febrile neutropenia (Jones et al., 2008) which can be very serious, leading to rapidl;y progressing infections which can possibly impact mortality. Preventative administration of anti-infectives is therefore suggested and if febrile neutropenia does develop, immediate action must be taken to avoid progression to sepsis (Freifeld et al., 2011). Figure 4 displays common infections following AutoHCT.
Allogenic HSC Treatment (AlloHCT)
AlloHCT uses HSCs collected from a disease-free Human Leukocyte Antigen (HLA) matched donor. These cells are transfused following high-dose chemotherapy and total body radiation. Due to additional complications in this therapy, research is limited. Figure 5 shows a diagrammatic comparison between the procedures involved in allogenic and autologous therapies.
AlloHST is the preferred treatment for a range of haematological diseases from leukaemia (Mapara et al., 2003), to myelodysplasia (O'Donnell et al., 1995) and has shown to result in restoration of haematopoiesis in these patients. This is thought to do with graft-versus-disease (GVD) effect which is thought to occur as a result of donor’s T cell launching an immune response against the patient’s disease (Khan, Agarwal and Agrawal, 2004).
AlloHCT may represent the only curative treatment for MM and shows greatest outcomes for those with a low tumour burden who enter remission following an initial chemotherapy treatment (Rajkumar, Kyle and Connor, 2018). However, even in individuals who respond well, relapse can occur. In a study on 80 chemotherapy-resistant MM patients receiving AlloHCT, just 5 had no signs of disease following the complete AlloHCT regimen (Bensinger et al., 1996) although, the nature of their disease severity may well have impacted on the significance of these findings.
DE and graft failure are important complications and can occur as a result of interactions between the patient’s own antibodies, natural Killer cells or T lymphocytes to the donor blood tissues. Most often this can be a failure of the donor cells to successfully graft, or a loss of these donor cells following a period of successful engraftment. The loss of donor cells can be due to autologous recovery which could confer long-term survival (Piccin et al., 2009) or more problematically, can precede the development of bone marrow aplasia or pancytopenia (Mattsson, Ringdén and Storb, 2008).
Extensive investigations of AlloHCT for MM patients have been limited as patients are usually randomised to a treatment depending on whether they have a potential HLA-matched donor where those without a donor undergo observation or a maintenance therapy. Results are analysed under the intention-to-treat (ITT) principle, as either donor or non-donor treatment groups. These have, perhaps unsurprisingly, been mixed. Although they have shown reduced relapse incidence and improved mortality, these findings do not always translate into a clear survival benefit. Furthermore, the quality of relapse treatments have significantly improved since many of these, earlier trials were completed, and in trials where novel chemotherapeutic agents such as, bortezomib are administered following relapse, demonstrate no survival benefit with AlloHCT (Rajkumar, Kyle and Connor, 2018).
Incidence of treatment-related mortality is significant and has shown to be upward of 50% in some studies mainly due to the myeloablative induction therapy, associated infections and GVHD (Shaw et al., 2003). The high mortality rate limits the scope of AlloHCT for MM. In fact, just 5% of MM patients can are thought to be viable candidates for this therapy as a result (Diagnosis and management of multiple myeloma, 2001).
While non-myeloablative treatments are emerging, the advent of newer chemotherapeutic agents and the improving efficacy of AutoHCT make these mortality improvements less notable. This raises questions about the overall significance and relevance of AlloHCT in MM treatment.
Use of cord blood
Cord blood transplants provide an interesting avenue for patients as tissue typing requirements are more relaxed (Gluckman et al., 1997). However, DE and graft failure have been shown to be more likely when compared to therapies using bone marrow-derived SCs, which has been known for some time (Rocha, 2001). Graft failure is especially likely in non-matched HLA grafts. Donor lymphocyte infusions (DLI) have been used in relapsing chronic myeloid leukaemia patients and could provide a solution to rejection and patients with decreasing donor T cell chimerism (Carlens et al., 2001). This must be balanced with common adverse effects of GVHD and more occasionally, bone marrow aplasia (Mattsson, Ringdén and Storb, 2008).
T Cell depleted AlloHCT
The use of T cell-depleted SCs may improve the incidence of GVHD and confer a benefit to mortality rates. However, this comes at the cost of efficacy and has been associated with increased disease relapse rate which has been linked to a reduction in the graft-versus-myeloma (GVM) effect (D'Sa et al., 2003).
T cell-depleted AlloHCT combined with DLI, has demonstrated progression-free survival of up to 65%, however significant treatment-related toxicities limit the scope of patients able to receive such therapy (Alyea et al., 2001).
Myeloablative induction is not routine (Lokhorst et al., 2010) and is often only considered for younger MM patients that are classed as having a high relapse risk, have identified an HLA-matched donor and understand and accept the associated high mortality (Rajkumar, Kyle and Connor, 2018).
Non-myeloablative induction therapies rely on the donor’s cellular immunity. They have shown improved treatment-related toxicity and mortality rates as low at 11% (Lokhorst et al., 2010) However, this comes at a cost to efficacy and the higher relapse rates seen when compared to myeloablative AlloHCT (Kumar et al., 2011).
AutoHCT followed by non-myeloablative AlloHCT
The GVD effect is clinically significant and represents a good outcome, it is associated with improved survival, especially in cases where chronic GVHD has occurred (Crawley, 2005). As with myeloablative AlloHCT, the best outcomes are seen with early intervention. As such, a number of treatment protocols recommend AutoHCT followed by non-myeloablative AlloHCT (Bruno et al., 2007).
A meta-analysis analysing over 1800 patients over 6 prospective trials, compared a double AutoHCT against a single AutoHCT followed by non-myeloablative AlloHCT. And found the AlloHCT had a 3.3 greater relative risk of treatment-related mortality with a comparable OS both before and after 36 months (Armeson, Hill and Costa, 2012).
Despite the main clear advantage of AlloHCT over AutoHCT is that it is potentially curative, alongside additional advantages of allogenic grafts being uncontaminated with disease or MGUS/SMM cells and the additional, important treatment benefit of the GVM effect (Mehta and Singhal, 1998), it is not often utilised. Primarily, due to toxicities, treatment-related mortality, and the increase in GVHD. The adverse effects are not balanced by a clear increase if efficacy over AutoHCT and chemotherapy (Barlogie et al., 2006). Furthermore, the ineligibility of many patients based on age or co-morbidities limits the scope of AlloHCT as the gold-standard MM treatment protocol.
Non-myeloablative AlloHCT may provide a solution but, this fundamentally relies on the GVM effect from the donor and is often associated with increase in GVHD. As such, overall outcomes from this are inconclusive and equally suffer from high treatment-related mortality, and toxicity (Krishnan et al., 2011). To this end a treatment of chemotherapy followed by AutoHCT is the gold-standard treatment for patients with MM, Older patients may require altered induction therapy prior to AutoHCT due to adverse effects of the chemotherapy (Rajkumar et al., 2020) despite almost all patients experiencing disease relapse and requirement for maintenance therapy.
The general patient population tend to be in their advanced years and therefore, toxicity, mortality and quality of life must all be carefully balanced. Ultimately, the decision to perform any type of HCT must largely be driven by patient values, preferences, and mitigation of harm. The pursuit of treatments which simply demonstrate great clinical efficacy is not always the most desired course of treatment for patients.