Is the IMWG frailty index an adequate tool for defining transplant eligibility in elderly patients with multiple myeloma?

Autologous stem cell transplantation (auto-SCT) is standard of care for the majority of patients with multiple myeloma (MM), although patients over the age of 65 are often not considered eligible for transplant. However, with the improvement in efficacy and tolerability of novel approaches in frontline treatments, the application of auto-SCT may expand amongst elderly patients with MM.

There are a number of comprehensive geriatric assessments (CGAs) that have been designed to optimize dosing regimens for elderly and comorbid patients with MM. Currently employed CGAs include, but are not limited to, the International Myeloma Working Group (IMWG) frailty score, Charlson comorbidity index (CCI), and hematopoietic cell transplantation comorbidity index (HCT-CI).

Effective and regimented selection criteria for the identification of patients able to withstand, and possibly benefit from, auto-SCT are yet to be defined. This article presents a summary of a study by Angelo Belotti and colleagues who investigated the influence of age on progression-free survival (PFS) in patients aged 65–75 with newly diagnosed MM (NDMM), by transplant eligibility. The study evaluated the IMWG frailty score, CCI, and HCT-CI as predictive prognostic tools, and aimed to verify the potential benefit of predicting treatment tolerability using a more objective CGA.¹

Study design

• Elderly patients (65–75 years) being considered for transplant for the treatment of NDMM (n = 131) underwent a CGA at the ASST Civil Hospital of Brescia, IT
• CGAs included the IMWG frailty score, CCI, HCT-CI, and the assessment of comorbidities using activity of daily living (ADL) and instrumental activity of daily living (iADL)
• Treatment intensity (auto-SCT vs no auto-SCT) was always decided by the attending physician, regardless of the CGA results
• Primary outcome: PFS
• Multivariate analysis was carried out across patient treatment cohorts, excluding patients classified as FRAIL patients, to evaluate the impact of disease-specific prognostic factors on patient outcome

Results

• Of the 131 patients considered for auto-SCT (Table 1), 85 (65%) were deemed eligible for transplant and comprised the intention to treat (ITT) study arm
• The ITT cohort received bortezomib-based induction:
  • bortezomib plus thalidomide plus dexamethasone (VTD; 94%)
  • bortezomib plus dexamethasone (VD; 5%)
  • bortezomib plus cyclophosphamide plus dexamethasone (VCD; 1%)
• In the ITT cohort (n = 85):
  • 72 patients (85%) proceeded to receive auto-SCT (70% single, 30% double)
  • 13 patients (15%) were subsequently considered ineligible for auto-SCT, reasons included progressive disease, severe renal impairment, and severe adverse events (SAEs) during induction therapy
• Forty-six patients (35%) were considered ineligible for auto-SCT and received reduced intensity first-line treatment:
  • bortezomib plus melphalan plus prednisolone (VMP; 89%)
  • daratumumab plus VMP (4%)
  • melphalan plus prednisolone (MP; 2%)
  • steroids/palliation (5%)

Table 1. Baseline patient characteristics by treatment intensity and IMWG frailty assessment

Patient response

• Transplant-related mortality was 0% across the entire study
• The proportion of patients achieving complete remission (CR) following frontline treatment was significantly higher in the ITT arm vs no auto-SCT group (Table 2)
• In the ITT group, the 2-year death rate was 13%,
  • causes of death were progressive disease (n = 10) and aspiration pneumonia (n = 1)
• At a median follow-up of 27 months, the PFS was significantly superior in the ITT arm vs the no auto-SCT group (35.6 months vs 9 months; hazard ratio [HR] = 0.42, [95% CI, 0.25–0.71], p = 0.013)

Table 2. Patient responses by auto-SCT treatment cohort

HCT-CI and CCI

• PFS was significantly superior in:
  • patients with an HCT-CI of 0–1 vs ≥ 2 (not reported vs 3 months; HR = 0.45 [95% CI, 0.23–0.91], p = 0.025)
  • patients with an HCT-CI ≥ 2 in the ITT arm vs no auto-SCT group (median 34 months vs9 months; HR = 0.5 [95% CI, 0.30–0.84], p = 0.008)
• Patient outcomes were not impacted significantly by CCI (0–3 and > 3):
  • PFS was higher in patients of all CCI scores in the ITT arm vs no auto-SCT group (median 51.4 months vs 16.9 months; HR = 0.37 [95% CI, 0.15–95], p = 0.04)

 IMWG frailty score

• No patients classified as FRAIL were put forward for auto-SCT (Table 1)
• In contrast to HCT-CI and CCI scoring, patient classification by IMWG score was evenly distributed
• Patient PFS was significantly superior in:
  • patients deemed FIT and UNFIT vs those considered FRAIL by IMWG score (median 32.9- and 29.6-months vs 9 months)
    • FIT vs FRAIL: HR = 0.02, (95% CI, 0.004–008), p < 0.0001
    • UNFIT vs FRAIL: HR = 0.03 (95% CI, 0.007–12), p < 0.001
  • patients deemed FIT and UNFIT in the ITT arm vs no auto-SCT group (median 35.6 months vs 8 months, HR = 0.54 [95% CI, 0.31–0.97], p = 0.04)
  • patients in the ITT arm aged 65–69 years vs ≥ 70 years (51.5 months vs 7 months, HR = 0.34 [95% CI, 0.17–0.7], p = 0.00370
  • UNFIT patients aged 65–69 years in the ITT arm vs no auto-SCT group (43.3 months vs 4 months, HR = 0.03 [95% CI, 0.003–0.24], p = 0.01)
• No significant differences in patient outcome were observed between
  • FIT and UNFIT patients
  • UNFIT patients aged ≥ 70 years in the ITT arm vs no auto-SCT group
• Multivariate analysis uncovered further factors that independently influenced PFS
  • ISS III, HR = 0.48 (95% CI, 27–0.84), p = 0.011
  • patient status (ADL ≤ 4 and/or iADL ≤ 5), HR = 0.41 (95% CI,21–0.77), p = 0.006

Conclusion

The study established that auto-SCT is a viable option, and clinically beneficial for patients with MM aged up to 75 years old and selected with comorbidity indexes. The majority of patients included were deemed eligible for auto-SCT, which was generally well tolerated.

At the clinical cutoff points employed in this study, HCT-CI (≥ 2) and CCI (> 3) proved of little benefit in defining auto-SCT patient cohorts, with patients in the ITT group eliciting a superior response to therapy irrespective of both CGA result and age. Patients considered FRAIL using the IMWG frailty system demonstrated considerably poorer outcomes than other patient subgroups, though all patients regarded as FRAIL were also considered ineligible for auto-SCT by the respective attending physician. This elucidates that the IMWG frailty score coincide with clinical observation in these patients. However, the IMWG frailty score was successful in predicting favorable PFS rates in patients classified as FIT (any age) and UNFIT (aged 65–69 years), and the authors support the use of this tool to better select auto-SCT candidates, especially in patients aged ≥ 70 years old. The study also identified disease-specific, independent prognostic factors for PFS.

Limitations

• Some subgroups were underrepresented
• Lack of application of alternative geriatric assessment methods
• Comorbidities are accounted for in the IMWG frailty score; however, other geriatric fields and their effect on transplant eligibility are yet to be investigated:
  • cognition
  • psychosocial status
  • polypharmacy
  • depression
• Novel drugs are becoming progressively available, which may narrow the gap between auto-SCT and non-intensive treatment success in NDMM

All things considered, the global improvement in basal fitness of aged patients with MM, the advancing efficacy and safety of frontline treatments, and the current expertise in auto-SCT, mean that older patients with MM will be eligible for auto-SCT. In that setting, and according to this independent validation study, CGA such as the IMWG frailty score could be relevant not only as a prognostic factor, but also in guiding treatment decisions for subgroups of elderly patients with MM.

Further resources

The MM Hub has reported on several studies investigating the value of such CGAs, which have demonstrated superiority over age alone when considering prognostic value. For a summary of the XVII International Myeloma Workshop (2019) session, ‘Management of Elderly Patients with MM’, which reviewed current treatment options and classification systems for frail patients with MM, click here.

A modified frailty index was employed in a subgroup analysis of the A.R.R.O.W trial (NCT02412878), investigating the relationship between frailty and PFS in patients receiving once-weekly vs twice-weekly carfilzomib in combination with dexamethasone for the treatment of relapsed/refractory MM, read a summary here.