Since therapy decision-making is ideally supported by FA to individualize treatment especially for frail patients, we integrated the R-MCI into our MM-TB.9 This integration allows all users to readily distinguish fit from frail patients and expeditiously adjust therapy intensity during our weekly performed interdisciplinary MM-TB discussion.9,10,13,14 While various functional tests, including different comorbidity scores and geriatric functional tests have been examined,13,27–30 the R-MCI was explicitly helpful to individualize treatment decisions and improving the tolerance of MM therapy.10 Additionally, the R-MCI has been recognized as the only comorbidity index that did not show significant differences in risk group distribution for both retrospective and prospective data, thus it was reliably assessable from both data sets.14 Another convenience is the user-friendly R-MCI homepage (www.myelomacomorbidityindex.org15).
While fitness assessments are used for tailoring therapy in other hematological diseases, these are not routinely established in MM patients, despite studies aiming to elucidate their usefulness.10,13,31 FA have been described as challenging to integrate into everyday clinics due to time constraints.13,32–34 The pivotal outcome of our study was the successful integration of the R-MCI into MM-TB protocols in > 90%, confirming previous studies of the R-MCI being resourceful.7–10,13,14,24,33 The establishment of the R-MCI in our MM-TB exemplified that frailty scores can be readily used in everyday clinics. Albeit the R-MCI calculation is specifically requested in our TOS MM-TB system, its use and integration for each patient therein was not obligatory within TOS during the assessment period, therefore the availability in > 90% of our large MM-TB cohort (n = 691 MM-TB cases; Fig. 1) was a success. Of note, the R-MCI was unavailable in 43 patients (6%) only, who had been externally referred to our UKF/CCCF: either missing data were reasons for its non-use, or physicians from different disciplines/departments were overriding the TOS-integrated R-MCI scoring. In these patients, the subjective assessment of the introducing physicians within the MM-TB sessions was utilized (which would have otherwise been complemented by the R-MCI). Meanwhile, we have improved the TOS MM-TB software to make it mandatory to enter the R-MCI parameters required for its calculation and thereby ensuring that R-MCI scores are available in 100% of patients. Consistent with previous analyses, we affirmed the rarity (2%) of frailty in < 60-year-old patients but acknowledged its significance. This was indeed important to determine, because in the only frail 53-year-old patient, CD38-based induction and ASCT had been the TB-recommendation, whereas CD38-based therapy alone would have been the better TB-choice according to the R-MCI assessment. We observed an increase in frailty to 36% in ≥ 70-year-old patients, albeit among them also 9% were fit and 55% were intermediate-fit. Dose reduction was performed in alignment with fitness levels in 38% in fit, 61% in intermediate-fit and 74% in frail patients. Decision-making without an available R-MCI revealed that in three frail patients, therapeutic decision could have been facilitated with exact knowledge of patients’ fitness. Consequently, therapy choices would have differed, if the R-MCI had been assessed before therapy initiation.
Of note in this study and different to others10,13 was, that our follow-up assessment of potential R-MCI changes was conducted earlier, after a median of 5 months, and that the R-MCI showed fitness improvement or stability in 55%, while a decline was observed in 45% of cases. In previous studies, where the follow-up was conducted after a median of 11 months, the R-MCI showed rather an improvement in 90% of patients than decline in only 10%.10,13 This indicates that it may be more advisable to evaluate changes in fitness and quality of life (QoL) after a longer period following the initiation of therapy, in order to prevent temporary therapy side effects from affecting QoL.10,13,21,35 Therefore, to possibly better capture MM patients’ advances in performance, therapy endurance and QoL, a defined latency of approximately 1 year appears more suitable than < 6 months. Likewise, the common practice of repeatedly assessing QoL domains (i.e. in QoL questionnaires in clinical studies) over very short periods may be less practical in view of our data.10,13,21,35
Our assessment of the MM-TB over a 3-year period revealed that 38% of patients were presented at least twice, resulting in at total of 691 MM-TB presentations. Among these patients, 71% had 2–3 MM-TB presentations within our observation period, while 29% had ≥ 4 presentations, and 9% had ≥ 6, supporting our view on very frequent TBs.19 MM-TB outcome data of 2020/21 − 9 and 2012-2014-analyses16 as well as our data here seem valuable, as we had previously described the postulated survival benefit in patients with ≥ 3 TB discussions as error-prone, occurring due to an immortal time bias (since patients need to survive long enough to be discussed more often). Therefore, time-biased results should not lead to the conclusion that more TBs will increase patients' survival. Instead, insightful discussions within one or few meaningful, long-lasting TBs, ideally in interdisciplinary teams, will generate most profound results for cancer patients.19,36
Of interest was, that the number of dose-reduced therapies in our entire cohort was higher than in previous studies (55% vs 41%10, respectively). This difference was likely related to our focus on MM-TB-patients, whereas Holler et al. had assessed physician-based therapeutic decisions of consecutive MM patients.10 Our results revealed that dose adaptions were regularly performed in frail patients. Although age-based dose reduction was also observed, they seemed more error-prone than if the R-MCI was included in the decision-making: older patients (≥ 70 years) were treated with dose reduction in 70%, albeit only 36% of them were frail, while 9% were fit and 55% were intermediate-fit. Consequently, this group of patients was rather undertreated in substantial numbers, as previously described.8 In line, younger patients can be frail and are important to decipher likewise. Therefore, the assumption that younger patients should always receive full-dose treatment, and elderly patients should not, may often be error-prone. This confirms that age alone is not sufficient to determine patients’ health status and therapy endurance. Instead, FA is superior in specifying patients’ constitution and biological age7–14,24 which is why treatment decisions should rather depend on FA tools, such as R-MCI or others.12,23,24,27–30
Strengths of our analysis included the precise examination of a large MM-TB cohort, the repeated R-MCI analysis, and the observed dose reduction, well-associated with R-MCI-, rather than age-subgroups. Our observation period of 3 years was substantial, and the detailed examination of patients with various numbers of TB presentations and deterioration in their health status (R-MCI) exhibited their complexity. This is in line with a prior analysis in MM patients with first- vs. later lines of therapy as compared to BSC,21 which, however, did not assess QoL changes - as here - in consecutive but different follow-up cohorts. Our analysis was well comparable to prior analyses9,16,19 and revealed the quality of one out of 24 TBs at our UKF/CCCF: while the quality results of our MM-TB, the decisions, pathway- and guideline-adherence, TB-compliance, referrer satisfaction, and improvement of clinical trial inclusion have been described previously,9,16,19 our data here proved that FA integrated into TBs is feasible and does support therapy decisions.
Limitations of the study were the single-institution approach and the range of patients’ ages (38–86 years), with 64% of patients being < 70 years old. This is, however, typical for MM patients in tertiary centers and suggests that our data are even more relevant in older patients. Additionally, the specific conditions for our study inclusion (MM-TB presentations ≥ 2, subsequent therapy being instituted, and performed follow-up R-MCI) focused on more complex MM patients (and a subset of our initial study cohort, Fig. 1). Our cohort did include patients receiving different treatment with induction, maintenance and later-line treatment, since our aim was not survival analyses, rather than to assess the completeness of the R-MCI within our MM-TB, whether therapy decisions in terms of fitness vs. age cohorts were different and how dose-reductions were performed. The aim of this study was therefore not to determine exact therapy choices of newly diagnosed MM vs. relapsed/refractory MM, or within different therapy lines, nor whether patients did profit from dose-reductions, because we and others had shown this in prior publications.7,8,10,27–30,37–44 Moreover, our follow-up assessment was conducted within a relatively short period which should ideally be performed after ~ 1 year, according to Scheubeck and Holler.10,13 Another criticism may arise from the retrospective evaluation of three patients (whose R-MCI was unavailable in TOS and who received different, more intensive treatment due to the unavailability of their R-MCI). This introduced bias as their outcome was known during our reassessment of therapy choices. Last, since we had examined survival repeatedly in similar MM cohorts, via R-MCI and age subgroups and in different MM-TB cohorts,7–10,13,14,16 this was not repeated here.
In conclusion, our results demonstrate the widespread use of the R-MCI within the MM-TB. Further research through prospective clinical trials seems essential to determine optimal, personalized treatment options for each patient. Building on this approach, Mian et al. published a systematic review including 43 clinical trials considering frailty tools and showed an encouraging trend to incorporate frailty assessments in clinical evaluations and treatment decisions,41 in line with our data of the R-MCI-integration in TOS-MM-TBs in > 90%. The association between the R-MCI and chosen therapy intensity was better than via age cohorts, which further underlines that the R-MCI is a more precise predictor than age alone. We could show that the recommendations to establish therapy decisions on FA can be directly implemented in TBs. Therapy decisions for intermediate-fit patients appear more complex than for fit and frail patients, because some intermediate-fit patients may endure full-dose treatment, while others need dose-reduction. Therefore, this group of patients should be analyzed further. Today, some MM experts distinguish only two groups of fit vs. frail patients.6,10,38 Prospective studies using the R-MCI as an important tool for therapeutic decision-making are in process at our CCCF. Most importantly, these and other important TB analyses have led to our better interpretation of cancer care, in close collaboration with statisticians16,22,45–47, which is essential to produce reliable evidence for future progress. We are grateful that these productive collaborations continue to exist at our and other CCCs.