Guidelines on the diagnosis, investigation and initial treatment of myeloma: a British Society for Haematology/UK Myeloma Forum Guideline
2021; Wiley; Volume: 193; Issue: 2 Linguagem: Inglês
10.1111/bjh.17410
ISSN1365-2141
AutoresJonathan Sive, Kirsty Cuthill, Hannah Hunter, Majid Kazmi, Guy Pratt, Dean Smith,
Tópico(s)Hematological disorders and diagnostics
ResumoThe objective of this guideline is to provide healthcare professionals with clear guidance on the anti-myeloma management of patients with newly diagnosed multiple myeloma. In all cases, individual patient circumstances may dictate an alternative approach. This guideline was compiled according to the BSH process at https://b-s-h.org.uk/media/16732/bsh-guidance-development-process-dec-5-18.pdf. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) nomenclature was used to evaluate levels of evidence and to assess the strength of recommendations. The GRADE criteria can be found at http://www.gradeworkinggroup.org. Recommendations are based on a review of the literature using Medline, PubMed, Embase, Central, Web of Science searches from the beginning of 2013 up to July 2019. The following search terms were used: myeloma; plasma cell leukaemia; AND risk; prognosis; cytogenetics; FISH; PCR; molecular; imaging; response; residual disease OR [chemotherapy; autologous; autograft; HDT/ASCT; allogeneic; allograft; stem cell; bone marrow; cord blood; haploidentical; tandem transplant; bortezomib; carfilzomib; ixazomib; melphalan; thalidomide; lenalidomide; pomalidomide; cyclophosphamide; dexamethasone; prednisolone; doxorubicin; bendamustine; immunotherapy; daratumumab; PDL1 inhibitor; CAR-T; frail; elderly; renal failure; renal impairment; kidney disease; maintenance; consolidation AND survival; outcome; relapse; progression; remission; response; residual disease; mortality; morbidity; side effects; adverse events; complication; neuropathy; thromboembolism; infection; quality of life; cost-effective] Review of the manuscript was performed by the British Society for Haematology (BSH) Guidelines Committee Haematology Oncology Task Force, the BSH Guidelines Committee and the Haematology Oncology sounding board of BSH. It was also on the members section of the BSH website for comment. It has also been reviewed by UK Charity Myeloma UK. These organisations do not necessarily approve or endorse the contents. Patients with suspected myeloma should be investigated using the tests listed in Table I. A bone marrow biopsy should be undertaken in patients in whom there is a clinical concern for end organ damage and/or those with a significantly elevated monoclonal protein (M-protein). The monoclonal protein should be quantified by densitometry of the monoclonal peak. Quantification of monoclonal immunoglobulin (Ig) A by electrophoresis can be complicated by migration into the beta region. International Myeloma Working Group (IMWG) guidance recommends that for IgA and IgD myelomas, quantitative immunoglobulin measurements are preferred.1 NICE guidance now recommends the use of serum free light chains (SFLC) rather than urinary Bence Jones protein (BJP), and studies have validated this.2 SFLC replaces BJP in these guidelines, although it is noted that BJP may still be required for some clinical trials. Urine albumin:creatinine ratio along with troponin and N-terminal pro-B-type natriuretic peptide (NT-proBNP) can be a useful screening tool for detecting amyloid. Skeletal survey has been replaced by cross-sectional imaging, including low-dose, whole-body computed tomography (CT), or ideally functional imaging such as computed tomography-positron emission tomography (CT-PET) or diffusion weighted whole body magnetic resonance imaging (MRI). Focal imaging (e.g., dedicated MRI scan of the spine and pelvis, or plain films of long bones) should be performed to look at specific sites in more detail if required. Imaging in myeloma is discussed in detail in recent UK and international guidelines.3, 4 All diagnoses should be reviewed at a multidisciplinary team (MDT) meeting. Myeloma should be diagnosed using the 2014 IMWG updated criteria.5 Table II shows the diagnostic criteria for myeloma, smouldering (asymptomatic) myeloma and monoclonal gammopathy of undetermined significance (MGUS). Table III summarises myeloma-defining events. The latest guidance reflects a number of changes compared to the previous 2003 criteria.6 End organ damage is no longer required to diagnose myeloma. Three biomarkers have been added to the myeloma-defining events, each of which is associated with an approximately 80% probability of the development of CRAB features (hypercalcaemia, renal impairment, anaemia and bone disease). These biomarkers (≥60% clonal plasma cells in the bone marrow, involved:uninvolved light chain ratio ≥100 and ≥2 focal lesions on MRI) are referred to as the SLiM criteria. Studies have shown the rate of progression to myeloma within 2 years is approximately 95%7, 8, 80%7, 9 and 70%,10, 11 respectively. Patients with a solitary focal lesion on MRI or equivocal findings should undergo interval imaging.4 The 2003 criteria did not specify the percentage of clonal bone marrow cells required for a diagnosis of symptomatic myeloma. Current guidance confirms 10% clonal plasma cells or biopsy-proven plasmacytoma is required. Current guidance also clarifies that the presence of osteolytic bone lesions >5 mm seen on CT or PET-CT (and not on skeletal radiography) is consistent with a myeloma-defining event. Increased uptake on PET-CT alone, without a corresponding lytic lesion, is insufficient to be a myeloma-defining event, but is associated with an increased risk of progression to myeloma.5 If there is doubt regarding equivocal or small lucencies (<5 mm), repeat imaging should be performed. Bone lesions should be biopsied if there are concerns they may represent bony metastases from concurrent malignancies. Osteoporosis with compression fractures is no longer a myeloma-defining event. Criteria regarding renal failure have changed, with creatinine clearance 500 mg/l is suggestive of cast nephropathy,12 so renal biopsy should be considered in those patients with SFLC <500 mg/l.5 Monoclonal proteins can cause other renal pathology (e.g., AL amyloidosis or monoclonal immunoglobulin deposition disease) in patients who do not meet the criteria for myeloma. Such cases are termed monoclonal gammopathy of renal significance.13 To be classed as a myeloma-defining event, CRAB events should be due to underlying myeloma and, if unclear, appropriate investigations should be performed to confirm this. Symptomatic hyperviscosity, amyloidosis and recurrent bacterial infections1 have been removed from the list of myeloma-defining events in the current guidelines, but may still require treatment. Diagnostic criteria for other related plasma cell dyscrasias, including solitary plasmacytoma with or without minimal bone marrow involvement, systemic AL amyloidosis and POEMS Syndrome can be found in the IMWG updated criteria for the diagnosis of myeloma.5 Cytogenetic analysis should be undertaken by interphase FISH (fluorescence in situ hybridization) on CD138-selected bone marrow cells. The bone marrow material used should be part of the first aspirate pull wherever possible.14 Table IV lists the cytogenetic abnormalities found to be of prognostic significance in newly diagnosed myeloma.15 Whilst t(11;14) is listed as standard risk, recent analysis of the Myeloma XI trial suggested patients with hyperdiploidy and no adverse lesions had superior outcomes compared to those with t(11;14).16 Early data suggests t(11;14) is a predictive biomarker for response to the BCL2 inhibitor venetoclax. t(4;14) is a poor risk marker, although the poor risk can be (partly) overcome by bortezomib-based therapy.17 The poor prognostic impact of del(1p) has mainly been described in patients treated with autologous stem cell transplantation.18 The number of extra copies of 1q is relevant; patients with amplification of 1q (four or more copies) having a poorer prognosis.16, 19 Del(13q) detected by FISH is no longer considered an independent prognostic factor.20-22 There is a lack of consensus internationally as to the percentage cut-off levels which should be used to signify a positive FISH result. The French group (IFM) suggested 60% was required for clinical significance in del(17p), although this has not been replicated in other studies, and sub-clonal TP53 copy number abnormalities have recently been shown to be associated with prognosis.23 The European Myeloma Network suggested 10% for translocations and 20% for copy number abnormalities.14 Less than 20% has been clearly associated with inferior outcome in the UK MRC Myeloma IX and XI trials. Smaller sub-clones may carry prognostic relevance, but data are currently limited. UK MRC Myeloma IX and XI studies found an association between the number of adverse cytogenetic lesions present and progressively shorter survival.16, 20 The International Staging System (ISS) defines three prognostic categories (Table V). The criteria reflect tumour burden and renal function (Beta-2 microglobulin) along with performance status (albumin). The ISS was initially developed in 2005.24 As such, the median overall survival (OS) associated with each stage (62 months vs. 44 months vs. 29 months) is out-dated. However, more recent studies have confirmed the prognostic significance of ISS in the era of novel agents25 and at relapse.26 The Revised-ISS (R-ISS) combines the traditional ISS with presence of high-risk cytogenetics (del(17p), t(4;14) or t(14;16)) or elevated serum lactate dehydrogenase (LDH).27 Data were pooled from 4,445 patients with newly diagnosed myeloma enrolled onto 11 international multicentre trials, 95% treated with novel agents. Three risk groups are defined, as shown in Table VI. In addition to the cytogenetic abnormalities discussed above, various recurrent genetic mutations have been associated with a poor prognosis in myeloma—for example, in CCND1, ATM and TP53.15, 28 The National Genomic Test Directory (https://www.england.nhs.uk/publication/national-genomic-test-directories) specifies the genomic tests commissioned for myeloma in the UK. Bi-allelic loss of TP53 [i.e., del(17p) plus TP53 mutation, seen in ~30% of del(17p)] has a significantly reduced OS.28 Gene Expression Profile signatures are also predictive of poor prognosis but are currently used only in the context of clinical trials.29 Plasma cell leukaemia (defined as 20% circulating plasma cells or a total plasma cell count in peripheral blood of at least 2 × 109/l) remains a poor prognostic factor,30 as does detection of low levels of circulating plasma cells by flow cytometry.31 Imaging studies can provide prognostic information; the presence and number of 18F fluorodeoxyglucose (FDG)-avid lesions on PET scanning at baseline and at response to treatment has the most data in this regard at the present time.32 Investigations should be based on the tests listed in Table I. (1C) Serum free light chain analysis should be used to investigate monoclonal light chains rather than urinary Bence Jones protein. (1B) Renal biopsy should be considered if SFLC 20% of cells should be considered significant. The significance of smaller clones is not clear. (2B) Revised ISS should be calculated on all newly diagnosed patients. (1A) Overall survival is the preferred outcome measure for assessing efficacy, using direct comparisons from Phase 3 trial data where possible. Progression-free survival (PFS) and response rate (RR) can be used as surrogate markers, although caution should be employed in their interpretation. Treatment cross-over in trials at progression means that a PFS advantage even in the absence of OS difference may still indicate a benefit from a treatment option; in this context, PFS2 (progression-free survival on the next line of treatment) can provide useful data.33 Increasingly, sustained Minimal Residual Disease (MRD) negativity is seen as a strong surrogate marker for long-term outcome.34 Myeloma predominantly affects an elderly population, many of whom are excluded from clinical trials; hence, there can be less certainty about the benefits of treatments and effects on quality of life in this group. Toxicities can be considerable, and dose modification is often necessary. Higher doses of corticosteroids35 and discontinuation due to adverse events36 are associated with worse overall survival in this population. Conversely, fitter older patients may receive inappropriate dose reductions if based solely on age. Evaluation of frailty was traditionally based on age and subjective clinician assessment. More recently, objective fitness scoring systems have been evaluated to estimate prognosis and guide dosing.37-41 The IMWG score is based on age, the Charlston Comorbidity Index and cognitive and physical conditions, while the UK Myeloma Risk Profile (UK-MRP) uses patient and disease factors.42 Prospective trial-based testing of these systems is ongoing, and consensus on their use has not yet been reached. As discussed below, autologous stem cell transplantation (ASCT) is recommended for younger, fitter patients. There is no formal definition of transplant eligibility and age alone is a poor indicator. Selected patients over the age of 70 may be suitable for ASCT with a low risk of mortality (3–5%). Transplant scoring systems can be used to assess fitness objectively and formal tests of cardiac, lung and renal function performed, although these are not currently standardised. A full discussion of side effects and dose reductions is beyond the scope of this guideline, but these have a significant bearing on drug choice and dosing. The Summary of Product Characteristic datasheets should be referred to. Patient preferences, including duration of therapy, and practical issues such as the need to travel to a day unit for parenteral treatments are important considerations, especially in the frailer patient population where quality of life as well as OS is important. Local familiarity with regimens can play an important role. Licensing and funding varies between countries and regions, and will change over time. The criteria for assessment of response continue to evolve and are defined based on paraprotein, bone marrow and imaging responses as: Stringent Complete Response (sCR), Complete Response (CR), Very Good Partial Response (VGPR), Partial Response (PR), Minimal Response (MR), Stable Disease (SD) and Progressive Disease (PD), with the more recent inclusion of MRD-based assessments by flow cytometry or sequencing and imaging.1 Outside of a clinical trial, light chain assessments can be made by SFLC assay rather than urine BJP quantification.2 Future trials will explore using MRD and functional imaging responses, but these are not currently used to make routine treatment decisions. This section discusses choice of drug treatment for newly diagnosed myeloma patients. Treatment decisions should be made within an MDT context, and may involve supportive care, surgery and radiotherapy, although these areas are not covered within these guidelines. The aim of treatment for all patients is to maximise the depth and duration of response while minimising toxicity in order to lengthen survival, improve quality of life, alleviate symptoms and prevent further organ damage. Drug regimens referred to in the text are listed in Table VII. Proteasome inhibitors (PIs) act by altering the degradation of proteins essential for cell cycle and growth.43 The first in class, bortezomib, was originally given on an intravenous, biweekly schedule, but appears to be equally efficacious with reduced peripheral neuropathy when given weekly and subcutaneously.44-47 A biweekly schedule may be used as initial therapy to try to achieve rapid tumour control in highly proliferative disease or with cast nephropathy-induced acute kidney injury. Carfilzomib is a second-generation PI with irreversible proteasome binding which has significant efficacy but higher rates of cardiac toxicity.48 Carfilzomib is given intravenously, and dosing schedules vary: 70 mg/m2 once weekly is better tolerated, with improved efficacy compared to 27 mg/m2 twice a week in the relapsed setting,49 but optimal dosing remains to be determined in the frontline setting. The oral PI ixazomib has limited data in the first line setting at this time. Immunomodulatory drugs (IMiDs) are oral agents that cause myeloma cell apoptosis primarily by interaction with cereblon. Mechanisms of action include degradation of the transcription factors IKZF1 and IKZF3,50 and immune modulation.51 The first drug in class, thalidomide, shows clinical efficacy, but is associated with high rates of venous thromboembolism (VTE) when used in combination with corticosteroids,52, 53 as well as tremor, neuropathy and constipation. The newer agents, lenalidomide and pomalidomide, have a lower VTE risk and are better tolerated, but have a higher incidence of myelosuppression, often requiring growth factor support.54 All IMiDs require risk-stratification and prophylaxis for VTE, as well as a pregnancy-prevention programme due to their potential teratogenicity. These remain a key part of myeloma therapy, with oral dexamethasone and prednisolone being the two most widely used in UK practice. Steroid toxicity can be underestimated, and doses should be reviewed and reduced if possible with long-term use. There are emerging data indicating that steroids can be stopped once patients enter a maintenance phase of treatment with equivalent PFS and OS.55 Once weekly dosing rather than four-day blocks during initial therapy is associated with lower toxicity and mortality.35 Higher dose treatments may be given in patients presenting with highly proliferative disease or with cast nephropathy-induced acute kidney injury. Alkylating agents (e.g., melphalan, cyclophosphamide) may be used in combination with other agents. The doses used are suitable for outpatient regimens, although myelosuppression and mucositis can still occur. More potent cytotoxic chemotherapy combinations such as DT-PACE are sometimes used in aggressive disease. Monoclonal antibodies, particularly the anti-CD38 antibody daratumumab, deepen response in combination with both IMiD and PI-based chemotherapy and are likely to be adopted in frontline regimens. Toxicities are manageable, but include first dose infusion reactions, interference with blood grouping and interpretation of low-level IgG monoclonal proteins.56 Other agents, including isatuximab (anti-CD38) and elotuzumab (anti-SLAMF7), have limited data in the first line setting. In direct comparisons, PI-based induction regimens with bortezomib or carfilzomib give greater RR, PFS and, in some trials and meta-analyses, OS benefit compared to non-PI-based regimens.57-61 The majority of first line studies have used bortezomib in various combinations in both the transplant eligible (TE) (Table VIII)59, 60, 62-64 and non-transplant eligible (NTE) contexts (Table IX).57, 65-68 Carfilzomib has been tested widely in TE patients (Table X).69-73 At the time of writing, these data are predominantly in abstract form, with OS data not mature. However, response rates and PFS are all at least as good as with bortezomib-based regimens. Cardiac risks have been highlighted, but the safety profile is acceptable in a younger population. In contrast, data in NTE patients (KMP vs. VMP) have not shown an RR or PFS advantage over bortezomib.74 The addition of a third agent often deepens response, although this has not always translated into a survival advantage. The addition of an IMiD increases RR in both the TE and NTE setting, although in direct comparisons, the addition of thalidomide to PI/steroid combination has not shown a survival advantage.68, 75, 76 Although VRD has not been directly compared to other bortezomib-based combinations in Phase 3 trials, a retrospective analysis did indicate a survival advantage with VRD over VCD or VD,77 and single-arm Phase 2 trial data show RR, PFS and OS at least as good as VTD, making this an attractive, well-tolerated option in both TE and NTE patients.57, 63, 78 As noted above, VRD is clearly superior to RD for PFS and OS in the large SWOG trial,57 and a reduced dose protocol (RVDlite) is well-tolerated in older patients, making this a preferred, well-tolerated treatment option.67 Alkylating agents are an alternative option for patients who cannot receive an IMiD. VCD can be used for TE patients, although RR is lower than with VTD.79 PAD is another effective combination that may be used in fitter patients, although its myelosuppressive nature is a drawback.64, 80, 81 Melphalan is contra-indicated in TE patients due to the risk of impaired stem cell harvest. In NTE populations, VCD and VMP are commonly used induction regimens, with the VISTA VMP schedule widely used, and showing a survival advantage over both MP82 and VTP.83 The UPFRONT study, however, showed equivalent benefit for VD alone compared to VMP or VTD, again reinforcing the importance of the PI/steroid backbone,68 as was shown in a similar Phase 2 study of VP, VCP and VMP.66 Daratumumab has been added to various induction regimens, demonstrating improved RR and PFS for NTE patients in combination with RD,84 and both PFS and OS with VMP,85, 86 and PFS for TE patients in combination with VTD.84, 87 OS data remain immature, but PFS data provide early evidence of benefit, and it is likely to be rapidly adopted into the frontline setting. Non-PI-based combinations are an option in frailer patients for whom an all oral regimen is preferred (Table XI). For patients without high-risk cytogenetic features, these may be more tolerable and therefore more beneficial for long-term use. In this context, RD has been shown to be more effective than MPT due to greater efficacy and better long-term tolerability.88, 89 The addition of an alkylating agent to an IMiD/steroid combination (e.g., CTD, CRD) can deepen responses,90, 91 although survival benefits have not been demonstrated. Lenalidomide-based combinations are generally preferred to thalidomide-based ones, with improved survival in CRD compared to CTD in TE patients,92 and similar responses but better tolerability shown with MPR versus MPT in NTE patients.93 For TE patients, treatment is continued to maximal response with minimal toxicity, generally between four and six cycles before harvest and ASCT. With lenalidomide-containing regimens, harvest should be performed after 4 cycles to prevent inadequate stem cell yield. For NTE patients, there is a move to continuous therapies—for example, with lenalidomide and daratumumab. This is based on improved PFS in the Myeloma XI94 and FIRST trials,95 although there remains uncertainty as to the benefit in terms of overall survival. The VMP regimen is, however, for a fixed duration, as per the VISTA trial. The aim should be treatment delivery and tolerability, using reduced doses if necessary. Patients achieving less than a PR may benefit from switching to an alternative schedule. The Myeloma XI trial data demonstrated that patients achieving less than a VGPR following CTD or CRD induction benefit from switching to a bortezomib-based regimen.96 However, most patients will now receive bortezomib as initial therapy. Many units use DT-PACE (with or without bortezomib) or similar regimens in fit patients to achieve a deeper response prior to transplant, although there is a lack of data in this area. As above, high-risk myeloma is defined by a number of factors, of which cytogenetics have the main impact on initial treatment selection. There is evidence that PI-based therapy may abrogate the risk of t(4;14) and 17p−, and should therefore be used in these patients if possible.80, 97 In older NTE patients, this should prompt the use of a PI-based regimen (e.g., VMP) above an oral non-PI combination (e.g., RD) where tolerated. This approach is supported by a pooled analysis of two separate trials of VMP and RD.98 For patients presenting with cast nephropathy-induced acute kidney injury, plasma cell leukemia or with a proliferative phenotype, biweekly bortezomib with high dose blocks of dexamethasone can be used for initial treatment. Intensive cytotoxic-containing regimens such as (V)DT-PACE are occasionally used for rapid debulking and for more aggressive presentation in younger patients. Treatment should be chosen according to individual patient factors to maximise the depth and duration of response while minimising toxicity, in order to lengthen survival, improve quality of life, alleviate symptoms and prevent further organ damage. (1A) Treatment combinations should be selected for individual patients based on efficacy, tolerability, transplant-eligibility, frailty, comorbidities, patient preference and local familiarity, as well as national and local licencing and payment criteria. (1A) Transplant-eligible (TE) patients should receive a PI (bortezomib or carfilzomib)/corticosteroid-based induction regimen. (1A) Triplet regimens deepen response and are generally recommended for TE patients with the addition of an IMiD (e.g. VRD, VTD, KRD) preferred to cyclophosphamide (e.g., VCD, KCD). (1A) For TE patients, the aim should be to achieve maximal response with typically four to six cycles of an induction regimen prior to consolidation with ASCT. Patients receiving a lenalidomide-containing induction regimen should receive a maximum four cycles prior to stem cell harvest. (1C) Melphalan should be avoided in TE patient due to concerns about reduced yield at stem cell harvest. (1C) For NTE patients, the aim should be to balance delivering tolerable treatment and minimising discontinuations whilst still using effective regimens. (1C) NTE patients may receive a PI or non-PI-based treatment regimen. Patients with high-risk cytogenetics should receive a bortezomib/corticosteroid-based regimen if possible. For others, a lenalidomide-based, non-PI containing regimen is also acceptable, and may be preferred for patient-based factors. (1B) For NTE patients, an alkylating agent (cyclophosphamide or melphalan) or IMiD (thalidomide or lenalidomide) agent may be added to a bortezomib/corticosteroid-based regimen. Lenalidomide is preferred to thalidomide. (2B) Frailty assessment, including the use of objective scoring systems, should be carried out for older and less fit patients. A multidisciplinary approach with input from care of the elderly specialists may be beneficial. (1B) Dose modifications should be considered for all frailer, less fit patients. (1A) For patients achieving less than a PR, an alternative regimen may be considered in order to deepen response. (2C) Daratumumab is well tolerated and improves response rates and survival. It can be added to combination regimens, as per licence. (2A) Bortezomib should normally be given subcutaneously on a weekly regimen. (1A) Patients with aggressive proliferative disease, plasma cell leukaemia or myeloma-induced cast nephropathy should receive biweekly bortezomib for initial treatment or, alternatively, a more aggressive combination schedule such as DT-PACE. (2C) Autologous stem cell transplantation following high dose chemotherapy has been standard of care for consolidation following induction treatment in those considered fit enough, since it was first demonstrated to prolong PFS and OS with acceptable low levels of transplant-related mortality (TRM).99 Subsequent randomised trials have shown improved response compared to chemotherapy alone.100-104 A majority of these have also shown improved PFS, and although a significant OS advantage was not demonstrated in all trials, this is likely related to variation in induction and consolidation therapies, and the use of salvage ASCT in those not receiving it up front. On balance, therefore, ASCT has demonstrated its efficacy as post-induction consolidation (Table XII). More recently, given the increases in the rates and depths of remission achieved with the introduction of novel agents given in combination and the toxicities of high dose chemotherapy, its place in the upfront management of myeloma has been questioned. Two recent trials have attempted to address the role of ASCT following modern induction combinations and whether it could be reserved for a later line of therapy. The IFM2009 trial randomised 700 patients following VRD induction to ASCT or extended VRD consolidation. All patients received lenalidomide maintenance for 1 year. Median PFS (50 months vs. 36 months), CR rate and MRD negativity were all significantly longer in the ASCT group, maintained across all risk groups. There was, however, no OS benefit at 4 years (81% vs. 82%).105 In the EMN02/H095 trial,106, 107 a comparison of ASCT (single or tandem) with VMP consolidation after VCD induction, demonstrated upfront ASCT was associated with improved PFS (64% vs. 57% at 3 years) but again not OS (85% vs. 85% at 3 years). These trials demonstrate that post-induction ASCT continues
Referência(s)