Treatment approach for the older, unfit patient with myeloma from diagnosis to relapse: perspectives of a European hematologist

Multiple myeloma (MM) is predominantly a disease of older adults, with a median age at diagnosis of 69 years. Approximately 1/3 of patients are aged >75 years at diagnosis, and about 10% are aged >85 years.¹  Age negatively affects survival^2,3 ; however, elderly patients are a heterogeneous population that varies significantly in fitness. Variations in fitness level may affect how well a given regimen is tolerated, especially the most recent and effective combination regimens, and they could influence clinical benefit.⁴  Older patients are also underrepresented in clinical trials; reasons for exclusion include predefined upper age limits, numerous comorbidities, and use of multiple medications. In a recent publication, the European Medicines Agency⁵  pointed out that “such frequent exclusion has generated a situation of ‘evidence biased,’ as opposed to evidence based medicine for older adults.” This selection is even more evident for the frail patients.

Recently, the International Myeloma Working Group (IMWG) introduced a scoring system to classify the fitness level of elderly patients based on age, comorbidities (Charlson Comorbidity Index), and patient-reported self-care and household management assessments using the Katz Activities of Daily Living (ADL) and the Lawton Instrumental Activities of Daily Living (IADL) scales, respectively.^4,6,7  The IMWG frailty scale classified patients as fit, intermediate, or frail; these classifications are able to predict survival and risk of toxicity from treatment in patients with newly diagnosed multiple myeloma (NDMM).⁴  The Revised Myeloma Comorbidity Index (R-MCI) provides another simple tool for assessing frailty and classifies patients into groups of low, intermediate, and high risk; it considers renal and lung functions, age, and Karnofsky index, but it also includes cytogenetics.^8,9  Finally, another study exploring frailty scoring considered the N-terminal fragment of the type B natriuretic peptide, finding it to be a useful predictor of survival.¹⁰  In the absence of ADL and IADL assessments, it is difficult to formally assess our patient, Mrs. A., using the IMWG frailty score, but she would likely be categorized as intermediate or frail. Her R-MCI score is 8 (9 being the worst score; www.myelomacomorbidityindex.org).

Apart from frailty score, it is important to note that diabetes and hypertension, as noted in Mrs. A., are common in elderly patients. A recent review of 1240 patients with comorbid multiple myeloma and diabetes suggested an adverse prognosis associated with the combination, especially with steroid-induced diabetes.¹¹ 

Our patient would need additional imaging if available. This is a rapidly evolving topic in myeloma. In our patient, imaging consisted of only whole-body X-ray (WBXR). Other techniques, such as whole-body low-dose computed tomography (WBLDCT), magnetic resonance imaging (MRI), and ¹⁸F-fluorodeoxyglucose positron emission and computed tomography (FDG-PET/CT), may be used if available. Although WBXR has been the standard of care for many years, it has several limitations. The guidelines of both the European Myeloma Network and the European Society of Medical Oncology have recently recommended WBLDCT as the modality of choice for the initial assessment of MM bone disease.^12,13  MRI, either axial (spine and pelvis) or whole body, is the gold standard for detection of bone marrow involvement and would be useful in our patient, who has multiple vertebral fractures, to rule out spinal or radicular compression. Experts have also recommended the routine use of FDG-PET/CT in place of WBXR. FDG-PET/CT has diagnostic value and is useful for detecting extramedullary disease. However, 10% to 15% of patients have disease that is not ¹⁸F-fluorodeoxyglucose avid, potentially leading to false negative scans. The value of positron emission and computed tomography to assess response, inform treatment decisions, and predict clinical outcome is better established in transplant-eligible patients than in our elderly, frail, high-risk patient. Finally, it should be emphasized that considerable heterogeneity now exists in clinical practice regarding imaging modalities in MM. In many countries, WBXR is currently the only available and affordable technique for assessing bone disease in a patient with NDMM (a recent review on bone disease in MM is in ref. 14).

With regard to prognosis, our patient is classified according to the Revised International Staging System (R-ISS) as stage II and harbors a del(17p) mutation in 80% of plasma cells.¹⁵  Del(17p), which includes loss of TP53, is present in 5% to 10% of patients with NDMM, and it is likely the genetic feature with the worst prognosis in MM.

The Intergroupe Francophone du Myélome (IFM) recently reported on outcomes in 1890 elderly patients (median age, 72 years; with 651 patients >75 years of age), including 1095 patients with updated data on treatment modalities and survival. Regardless of treatment (which included melphalan + prednisone [MP], bortezomib + melphalan + prednisone [VMP], melphalan + prednisone-thalidomide [MPT], and lenalidomide + dexamethasone), both t(4;14) and del(17p) were found to be associated with worse clinical outcomes. The median progression-free survival (PFS) rates in patients with t(4;14) and del(17p) were 14 and 11 months, respectively, compared with 24 months in patients without these features. Similarly, median overall survival (OS) rates were 32 and 19 months, respectively, compared with 50 months.¹⁶ 

According to her R-MCI score of 8, the OS of our patient is expected to be ∼2.5 years (www.myelomacomorbidityindex.org), and her PFS is expected to be about 1 year. However, these results were established in patients treated with MP, MPT, and VMP and may improve with the most recent regimens, such as lenalidomide + bortezomib + dexamethasone (RVd).

Finally, the respective prognostic values of frailty and poor cytogenetics have recently been investigated in the Myeloma XI MRC study.¹⁷  Investigators noted that performance status had an impact on outcome at all ages, suggesting that physical frailty was more important than age itself. With advancing age, cytogenetic risk also had a smaller influence, and R-ISS stage had a greater influence on outcome.

Overall, the patient is sick, frail, and difficult to treat, with high-risk myeloma. As such, she represents the myeloma population with the highest unmet medical need.

Regarding frontline treatment options, the 2017 European Society for Medical Oncology Clinical Practice Guidelines proposed VMP, lenalidomide + low-dose dexamethasone (Rd), and RVd as preferred options. Other options are MPT, cyclophosphamide + thalidomide + dexamethasone (CTD), cyclophosphamide + bortezomib + dexamethasone (VCD), or bendamustine + prednisone.¹² 

If access to the regimen is available, RVd likely offers the best clinical outcome for our patient. However, at the time of this report, it represents a standard option in only a few countries, including in Europe. RVd has been investigated in both younger and older patients.^18-20  Importantly for our patient, it has also been developed as a modified regimen (RVd lite) for transplant-ineligible patients, including some elderly patients, under the assumption that it could be tolerable by a significant proportion of older patients.²⁰  In the phase 2 RVd lite study, patients received 9 5-week cycles of RVd lite (15 mg oral lenalidomide daily on days 1-21; 1.3 mg/m² subcutaneous bortezomib once weekly on days 1, 8, 15, and 22; and low-dose dexamethasone), followed by 6 cycles of consolidation and lenalidomide maintenance until disease progression. In the admittedly limited subset of 47 patients who were evaluable for response, an encouraging 91% response rate was noted, including very good partial response or better in 70% of patients; the median PFS was 35 months. Only 2 patients discontinued treatment due to toxicity. However, dose reductions occurred in almost 80% of patients, showing that careful patient monitoring is imperative, especially in frail patients who are considered for this approach. The RVd regimen developed in the SWOG study has 3-week cycles and lenalidomide + dexamethasone maintenance, and it would be less tolerable in a frail patient.¹⁹  Duration of induction therapy was 8 cycles in the SWOG study and 9 cycles with RVd lite.^19,20 

In Mrs. A.’s case, the RVd lite regimen would be given on a 35-day cycle, with an initial dose of 15 mg lenalidomide every other day for 21 days (due to her renal impairment); 1.3 mg/m² bortezomib on days 1, 8, 15, and 22; and low-dose dexamethasone (40 mg/wk; reduced to 20 mg/wk if needed).

The use of RVd lite in our patient is also supported by the presence of del(17p). Indeed, the IMWG recently recommended that patients at high risk receive triplets containing a proteasome inhibitor and an immunomodulatory agent.²¹  However, the presence of del(17p) still seems to be a poor prognostic feature with RVd, although data are limited. In the IFM 2009 study in younger transplant-eligible patients, of the 259 patients with evaluable cytogenetics in the arm that received RVd alone (8 cycles followed by 1 year of lenalidomide maintenance), 15 (6%) patients were found to harbor a del(17p) mutation (defined as plasma cells >60% positive), and their median PFS and OS were 15 and 54 months, respectively (Loiseau A, unpublished data for cytogenetics).¹⁸  An ongoing study (EudraCT2017-002238-21) is evaluating RVd in combination with a CD38 monoclonal antibody (isatuximab) in transplant-ineligible patients. However, this 4-drug regimen would not be suitable for frail patients.

In our patient, Rd will not be optimal, although it would be a good option for a standard-risk patient. In the phase 3 FIRST study, the median PFS rates in high-risk patients were 8.4 and 17.5 months for Rd continuous and lenalidomide + low-dose dexamethasone 18 months (Rd18), respectively. The median OS rates in high-risk patients were 29.3 and 24.3 months for Rd and Rd18, respectively.²²  The study enrolled patients with severe renal impairment; 9% of patients had creatinine clearance (CrCl) < 30 mL/min, and in this subpopulation, independent of cytogenetics, 4-year PFS and OS with Rd were 22.2% and 41.6%, respectively.²³ 

Because only a few countries currently have access to RVd, other treatment options must be discussed, with consideration of patient access to these drugs. If bortezomib is available but not lenalidomide, bortezomib + dexamethasone (VD), VMP, or VCD would likely be used. Both melphalan and cyclophosphamide would need dose reductions, at least at the start of therapy, due to Mrs. A.’s CrCl of 15 mL/min. The VISTA VMP registration study did not enroll patients with severe renal impairment, which is frequently the case in registration studies.²⁴  The FIRST study considered a dose reduction of 50% for melphalan in patients in the MPT arm with CrCl < 30 mL/min; in our elderly, frail patient, a clinician may use similar dose reductions with other MP-based regimens as well as with cyclophosphamide to remain cautious at the start of therapy. Bortezomib would be given twice weekly for 1 or 2 cycles (also referred to as VISTA cycles if VMP is preferred). VD would be an attractive initial regimen in our patient’s case. Both bortezomib (twice weekly) and dexamethasone (possibly at an intermediate rather than a low dose for initial cycles) are rapidly active drugs and may offer our patient a chance for at least partial improvement in renal function. VD would likely be easier to manage than VMP or VCD at the start of therapy; this regimen may be transitioned into VMP or VCD after a few weeks if renal function improves and as tolerance allows. If only thalidomide is available, MPT or CTD might be used, with thalidomide given at 100 or 200 mg daily and low-dose dexamethasone given in the case of CTD (or at an intermediate dose for the first 1 or 2 cycles as discussed previously for VD).

In the near future, daratumumab-based regimens will be available. Daratumumab-VMP has already shown superiority over VMP, with a favorable safety profile (although patients with CrCl < 40 mL/min were not enrolled),²⁵  and daratumumab-Rd (NCT02252173) may become a new standard of care. Other Rd-based regimens, such as ixazomib (NCT01335399) or elotuzumab-Rd (NCT01850524), may also become standard. The daratumumab-VMP study also assessed minimal residual disease (MRD) and showed promising results, with 22% of patients achieving MRD negativity at 10⁻⁵. The regimen has drawn the attention of the myeloma community due to effects on MRD in transplant-ineligible patients, but results remain preliminary. At present, basing therapy on MRD in elderly frail patients is premature.

Mrs. A. opts for RVd lite (oral lenalidomide, subcutaneous bortezomib, and oral or intravenous dexamethasone). After 6 monthly cycles, reevaluation reveals the following: M protein by serum protein electrophoresis (SPEP), negative; immunoglobulin A (IgA), 0.1 g/dL; β-2–microglobulin, 1.4 μg/mL; creatinine, 1.2 mg/dL; free κ, 15 mg/L; free λ, 25 mg/L; free light chain ratio, 0.60 (normal); and urine protein electrophoresis, trace λ.

The RVd lite regimen has achieved very good partial response after 6 cycles, with a significant improvement in renal function. The transplant option may be discussed, but deciding whether to proceed to autologous stem cell transplant (ASCT) in this situation is not easy. In many countries, ASCT remains the standard of care for transplant-eligible patients, and transplant is considered in patients younger than 80 years, assuming that they are not frail.^26,27  The proportion of patients ≥65 years proceeding to ASCT has regularly increased over the past 20 years, and transplant-related mortality rates in patients 65 to 69 and ≥70 years of age are similar (2%-3%) to that in younger patients.²⁸  PFS and OS have also been shown to be statistically similar in patients <70 vs ≥70 years of age.²⁹  However, in the transplant-eligible population of the MRC Myeloma XI study, only 5% of patients were 70 to 75 years of age, and ASCT delivery was less common in this age subgroup (∼40%).²⁷ 

In Mrs. A.’s case, geriatric assessment at diagnosis but also before transplant would be helpful, albeit not definitive. Our 73-year-old patient has improved her renal function, but she still has some renal impairment (glomerular filtration rate by the Modification of Diet in Renal Disease equation is 47 mL/min, despite a creatinine level of 1.2 mg/dL). At diagnosis, she had a Karnofsky performance status of 70, and she has comorbidities as well as some degree of frailty. However, her poor cytogenetics with del(17p) would favor ASCT, with favorable results shown in younger transplant-eligible patients.³⁰  Moreover, many of the frailty and renal issues seem to be related to the disease, which may also support ASCT. A decision to proceed to ASCT should be made carefully after the best possible assessment of benefits and risks in collaboration with the patient, family members, and any other caregivers. Tolerance of the initial RVd cycles will be a key factor in the transplant decision. If proceeding to ASCT, melphalan dosing (standard 200 mg/m² or reduced dose 140 mg/m²) would be discussed. However, if the reduced dose is chosen, its benefit in the context of RVd induction followed by maintenance seems uncertain. My opinion would most likely not be in favor of ASCT for Mrs. A, because the procedure would be more acceptable for a fit, elderly patient. Finally, we may assume that new combination regimens, especially those combining an immunomodulatory drug, a proteasome inhibitor, and a CD38 monoclonal antibody, might make ASCT less attractive in the future, particularly for older patients with standard-risk disease.

After the first 6 cycles, Mrs. A. will receive some additional cycles of RVd if tolerated before moving on to consolidation and/or maintenance. Duration of induction therapy has been 8 and 9 cycles in the SWOG and RVd lite studies, respectively.^19,20  High-risk disease supports continuous therapy. Some patients will be able to stay on RVd until progressive myeloma, but it is anticipated that many patients will discontinue dexamethasone and bortezomib at some point.

Mrs. A. has an excellent response and opts to continue therapy for 6 more cycles: no M protein, IgA of 0.5 g/dL, and only trace stability with an M protein no longer detectable on SPEP and a negative serum immune electrophoresis; however, she does have grade 2 peripheral neuropathy.

Mrs. A. is now in complete remission but has grade 2 peripheral neuropathy. At this point, it is appropriate to continue with lenalidomide alone at 10 mg (or 15 mg) daily for 21 days of a 28-day cycle. Maintenance will be continued until intolerance or progression.

Apart from the initial antimyeloma regimen, Mrs. A. will need supportive therapy. Optimal use of bisphosphonates (BPs) should be discussed in our patient who has extensive bone disease and severe renal impairment (but no hypercalcemia). Renal insufficiency is an infrequent but major complication associated with BP therapy. Zoledronic acid should not be used in patients with severe renal impairment. In patients with CrCl < 30 mL/min, 90 mg pamidronate might be infused over 4 to 6 hours.¹⁴  However, in the absence of tumor-induced hypercalcemia at the time of diagnosis, we would consider that the risk outweighs the benefit, and therefore, we would not use pamidronate. In the case of our patient, we would start with antimyeloma therapy alone, adding zoledronic acid if her CrCl improves to >30 mL/min. If her CrCl improves to 30 to 60 mL/min, we would decrease the dosage of zoledronic acid according to the package insert. Daily supplementation of calcium and vitamin D is recommended. The duration of treatment and dosing interval are 2 other important issues that are still not fully resolved. Current guidelines recommend treating patients with BPs for ≤2 years and then considering a pause in treatment until relapse.³¹  With new active regimens, such as RVd, if patients achieve a very good or complete response, a physician may opt for a 1-year treatment with zoledronic acid given every 4 weeks, subsequently considering an additional year with zoledronic acid given every 12 weeks. However, dosing interval and duration of treatment remain at the discretion of the treating physician.

Our patient also seems to be at high risk of treatment-emergent infection at grade ≥3 in the first 4 months of therapy and therefore, might benefit from antibiotic prophylaxis.³²  In patients receiving bortezomib, the use of oral acyclovir or valacyclovir is also necessary as prophylaxis against herpes zoster. In patients receiving thalidomide or lenalidomide, antithrombotic prophylaxis (usually with aspirin) is indicated. Patients may also require analgesics, but they must be instructed to avoid nonsteroidal anti-inflammatory drugs as well as other potentially nephrotoxic drugs.

Finally, it is important to keep in mind that, for this patient who was already taking 3 different medications, myeloma treatment will mean the addition of several drugs, including antimyeloma medications, prophylactic treatments, analgesics, and other supportive drugs. Although the patient seems mentally fit, compliance with such a complex regimen could be a challenge and would require careful medical education, possibly help from relatives, frequent visits, and consistency and continuity of care.

Mrs. A. is given maintenance treatment with oral lenalidomide 15 mg daily for 21 days of a 28-day cycle. She remains in remission for 2.5 years but then develops new lytic bone lesions, with M protein measured at 0.5 g/dL.

Multiple considerations related to patient or disease characteristics, the efficacy and tolerance of previous regimens, and patient preference must be taken into account when selecting a treatment for relapse.

In a recent study investigating real-world outcomes in 4997 European patients, it was shown that the percentages of patients who reached second- and third-line treatments were only 61% and 38%, respectively.³³  These results are clearly not optimal, and they are likely even worse in elderly patients.

Treatment options at first relapse in an elderly, frail patient are strongly correlated to drug access and therefore, vary greatly across countries, ranging from palliative therapy to advanced treatments, such as pomalidomide and daratumumab. With unlimited access to drugs and RVd lite followed by lenalidomide maintenance for frontline treatment, pomalidomide + daratumumab + dexamethasone would likely be favored in the second line, but it would probably only be possible today in the United States. Pomalidomide + cyclophosphamide + dexamethasone is an alternative option. Pomalidomide + dexamethasone or daratumumab as monotherapy would achieve inferior results.

Proteasome inhibitor–based regimens are alternative options based on either carfilzomib (carfilzomib + dexamethasone or carfilzomib + cyclophosphamide + dexamethasone) or ixazomib (ixazomib + cyclophosphamide + dexamethasone). However, carfilzomib would need to be used cautiously, likely at a reduced dose, in this frail patient with hypertension.

Mrs. A. receives second-line pomalidomide + daratumumab + dexamethasone. What are her options at the time of progression?

Second relapse in our patient has limited options. It may benefit from rechallenge with a bortezomib-based regimen if peripheral neuropathy has resolved, ixazomib-based or carefully designed carfilzomib-based regimens, or inclusion in a clinical study if available. Reevaluation of cytogenetics could be helpful in search of a t(11;14) translocation, which may trigger the use of venetoclax.

It is important to emphasize that Mrs. A. did not receive prior alkylating agent. In fact, at this stage, an old traditional regimen based on melphalan or cyclophosphamide might be beneficial. An alkylating agent–free regimen for frontline treatment is fully acceptable and standard in some countries. However, the question remains of whether an alkylator should then be given in second or third line, because these agents have efficacy, are better used in earlier lines of therapy, and are affordable in all countries. Indeed, is it acceptable to die of myeloma without exposure to an alkylating agent when combinations with novel agents in gentle regimens for elderly patients may have a role in myeloma?

The case of Mrs. A. highlights the many challenges that we face when treating older patients with myeloma. A newly diagnosed, frail, elderly patient with del(17p) and CrCl of 15 mL/min is very difficult to treat and represents the patient group with the highest unmet medical need. In many countries, cytogenetics evaluation is not routinely performed, and older patients, especially those with poor clinical features such as severe renal impairment, may not be referred to expert myeloma centers. For these reasons and others, it remains difficult to estimate the size of this population in real-life settings. Also, their numbers are not easily assessed in the context of clinical studies, because the majority of these studies continue to exclude patients with severe renal insufficiency, also raising the issue of how well study results can be extrapolated to real life. To estimate the size of this patient group, we looked at the database of the FIRST study, which assessed cytogenetics, including the high-risk mutations t(4;14), t(14;16), and del(17p), and enrolled patients with CrCl < 30 mL/min). In a total of 762 patients evaluated for cytogenetics, only 16 (2%) patients had high-risk cytogenetics as well as CrCl < 30 mL/min, but real-world experience indicates that this percentage is an underestimation. This discrepancy points to the need for reliable real-world data.

Significant progress has been made in the past 15 years due to novel agents.³⁴  Additional new drugs and strategies are needed to improve the clinical outcome of our patient, Mrs. A. However, if we can achieve this goal in an elderly, frail patient, other patients with standard-risk myeloma and more favorable clinical features could experience even greater improvement in their prognosis, and some might be cured.

Thierry Facon, Department of Haematology, Centre Hospitalier Universitaire Lille, Hôpital Claude Huriez, rue Michel Polonovski, 59037 Lille Cedex, France; e-mail: thierry.facon@chru-lille.fr.