Figures
Abstract
Progression-free survival (PFS) is a common primary endpoint in newly diagnosed multiple myeloma (NDMM). Patients with NDMM typically have longer PFS and are more likely to achieve minimal residual disease (MRD) or complete response (CR) compared to patients with relapsed or refractory multiple myeloma. Response-based surrogate endpoints may hold value given the longer follow-up time required to evaluate PFS in NDMM. In this work, systematic literature reviews of Medline, Embase, and Cochrane databases (2010-06/2020) and relevant congresses (2018–2020) were performed to identify randomized clinical trials (RCTs) and real-world studies in NDMM reporting median PFS and objective response. Associations between PFS and each response endpoint were evaluated using Pearson’s product-moment correlation weighted by sample size in each RCT arm. Unadjusted and adjusted weighted linear regression models were applied to estimate the gain in median PFS associated with each response endpoint. Statistically significant correlations were identified for median PFS with overall response rate (ORR; Pearson r = 0.59), CR (r = 0.48), stringent CR (sCR; r = 0.68), and MRD (r = 0.69). The unadjusted models estimated 0.50 (95% CI: 0.36, 0.64; p<0.001), 0.42 (95% CI: 0.25, 0.58; p<0.001), 1.05 (95% CI: 0.58, 1.52; p<0.001), and 0.35 (95% CI: 0.12, 0.58; p = 0.006) months of median PFS gained per point of ORR, CR, sCR, and MRD, respectively. Associations for median PFS remained statistically significant in models adjusted for age and treatment type with ORR (0.35, 95% CI: 0.21, 0.49; p<0.001), and adjusted for age and International Staging System risk stage with CR (0.29, 95% CI: 0.16, 0.41; p<0.001). Due to small sample size, adjusted models could not be constructed for sCR or MRD. Nevertheless, evidence of significant survival benefit (p<0.05) associated with MRD negativity and sCR was identified across real-world studies. These findings provide support for the use of response outcomes as surrogate endpoints to estimate PFS benefit in NDMM.
Citation: Daniele P, Mamolo C, Cappelleri JC, Bell T, Neuhof A, Tremblay G, et al. (2022) Response rates and minimal residual disease outcomes as potential surrogates for progression-free survival in newly diagnosed multiple myeloma. PLoS ONE 17(5): e0267979. https://doi.org/10.1371/journal.pone.0267979
Editor: David Dingli, Mayo Clinic, UNITED STATES
Received: October 18, 2021; Accepted: April 19, 2022; Published: May 12, 2022
Copyright: © 2022 Daniele et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was funded by Pfizer. The following Pfizer employees were involved in the study design, decision to publish, and preparation of the manuscript: Joseph C. Cappelleri, Carla Mamolo and Alex Neuhof. Timothy Bell was an employee of Pfizer at the time of writing and served the same roles as above.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests:The following authors are Pfizer employees: Joseph C. Cappelleri, Carla Mamolo and Alex Neuhof. Timothy Bell was an employee of Pfizer at the time of writing. Patrick Daniele, Anna Forsythe, Mihaela Musat, and Gabriel Tremblay are employees of Purple Squirrel Economics, which was a paid consultant to Pfizer in connection with the development of this manuscript. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Multiple myeloma (MM) is the second most common hematologic malignancy, with approximately 32,270 new cases diagnosed each year in the US [1, 2]. In recent years, a range of options have been approved to treat patients with newly diagnosed multiple myeloma (NDMM) [3]. Proteasome inhibitors, immunomodulators, and monoclonal antibodies offer substantial improvements in remission rates and survival, particularly when used early in the disease course; however, the five-year survival rate of 55.6% between 2011 and 2017 suggests a substantial opportunity remains to improve patient outcomes [1, 2, 4–6]. While these gains are promising, they present challenges with respect to demonstrating the benefit of new therapeutics in a clinically relevant timeframe, since planned trials in the newly diagnosed patient population using survival endpoints now require much longer duration for mature data. Patients would benefit from timely access to efficacious treatments, if efficacy can be established earlier in the course of treatment, before lengthy follow-up.
Based on regulatory assessments, there is precedent for acceptance of response endpoints for accelerated approval pathways, but regulatory agencies expect additional confirmatory data from survival endpoints [7]. Likewise, payers and health technology assessment (HTA) agencies expect long-term outcome evidence for their assessments [8]. While overall survival (OS) is the gold standard in oncology trials, few studies in NDMM reach median OS at the time of primary endpoint publication or drug approval [4]. This is largely due to the efficacy of modern first-line treatments as well as the availability of numerous subsequent treatment options for multiple lines of life-prolonging therapy. In addition to the extended time for survival data maturation, additional lines of treatment and supportive care may confound the OS data from the initial trial [9, 10]. In the NDMM setting, median OS is over 9 years in several recent trials, including the CALGB 100104 study of lenalidomide maintenance therapy following autologous stem cell transplantation, which not only set a new benchmark but necessitated a meta-analysis of multiple phase 3 trials that employed progression-free survival (PFS) or time to progression (TTP) as the primary endpoint for regulatory approval [11, 12].
Because of design and timing challenges as well as the potential for confounding, recent studies have employed either PFS or, to a lesser extent, TTP in trials in the NDMM setting [4]. A meta-analysis intended to explore the relationship between PFS and OS among 21 MM randomized clinical trials (RCTs) demonstrated a moderate-to-strong correlation; however, recent exceptions have been noted [13, 14]. Regardless, PFS is the most common primary endpoint in contemporary clinical trials in the NDMM setting. Nonetheless, the sample size and follow-up requirements to demonstrate significant improvements present challenges that have increased interest in identifying viable surrogate endpoints, such as overall response rate (ORR), complete response (CR), stringent complete response (sCR), or minimal residual disease (MRD), all of which are clinically relevant but have not been completely explored in patients with NDMM [4]. Patients with NDMM are more likely to achieve negative MRD status or CR than patients with relapsed or refractory disease, in addition to longer PFS [15]. In an analysis of recent RCTs reporting both MRD and PFS in the NDMM setting, median PFS of included studies ranged from 18.1 months to 56.3 months; however, median PFS in newly diagnosed, MRD-negative patient populations as long as 110 months have been reported [16–18]. Coupled with the fact that MRD negativity is recognized as a strong prognostic factor for both OS and PFS, its inclusion in trials and use in the clinical setting to monitor patients and help inform treatment decisions is on the rise [15]. Response-related endpoints and MRD status allow for a more rapid assessment of drug activity in the population of interest and may serve as early efficacy indicators despite not measuring survival directly. Recent meta-analyses have supported the utility of MRD negativity as a surrogate for PFS and OS in patients with NDMM [18, 19].
We therefore investigated the role of response endpoints as surrogates for survival endpoints in NDMM RCTs, while also assessing the literature-reported association of MRD with survival in real-world studies.
Methods
Literature search and article selection
A systematic literature review (SLR) of Medical Literature Analysis and Retrieval System Online (MEDLINE®), Excerpta Medica Database (Embase®) and Cochrane collaboration was performed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to retrieve articles published between 2010 and the search date of 26 July 2020 [20]. Additionally, abstracts from relevant congresses from 2018 to 2020 as well as bibliographies of systematic reviews and meta-analyses were reviewed to ensure all relevant studies were captured. The SLR scope was defined in terms of the patient population, intervention, comparators, outcome measures, and study design (PICOS statement; S1 Table). The SLR search strategy is provided in S2 Table. Eligible records were English-language interventional studies reporting ORR, CR, sCR, and MRD, in addition to median PFS associated with induction therapy with regimens including only approved treatments (bortezomib, lenalidomide, carfilzomib, ixazomib, daratumumab, thalidomide, and/or melphalan) in the NDMM setting. RCTs were included regardless of blinding and single-arm studies were also included. Evidence from observational studies (both prospective and retrospective), case series, and case reports was excluded. From included studies, all reported endpoints, study designs, and patient characteristics were extracted and used for the statistical analysis. In addition, a separate SLR for real-world evidence (RWE) studies was conducted using similar search parameters and constraints to further investigate the association between MRD and survival.
Statistical analysis
Results of the SLR were used to construct analytic datasets based on individual study arms. A meta-analysis was then performed on the aggregate data to estimate correlations and construct regression models weighted by study sample sizes [21]. All reported endpoints were considered as possible surrogates for median PFS since median OS generally was not reached in NDMM studies. Correlations between median PFS and each surrogate endpoint were estimated using Pearson’s product–moment correlation weighted by sample size in each study arm. The strength of the association was then categorized based the recommendations of the Institute of Quality and Efficiency in Health Care: low correlation (r ≤ 0.7), medium strength correlation (0.7 < r < 0.85) and high correlation (r ≥ 0.85) [22]. Survival gains associated with improvement in each surrogate endpoint were estimated using the following weighted linear regression models:
- Unadjusted: Median PFS Months = β0 + β1Surrogate + e
- Adjusted: Median PFS Months = β0 + β1Surrogate + βiXi+ e
where β0 is the intercept, β1 is the coefficient per percentage point of each surrogate value (% ORR, and % CR), Xi and βi are observed values and coefficients for covariates, respectively.
Diagnostic plots were used to verify the assumptions of the linear regression models. Moreover, multivariable models were constructed to assess covariates for adjustment, including key prognostic patient characteristics such as median age, sex, eligibility for transplant (ineligible only vs eligible and ineligible), International Staging System (ISS) risk stage 3 percent, and high cytogenetic risk. Trial design (RCT vs single arm), treatment phase (induction + maintenance vs induction only), and treatment type were also included. Due to the similarity of outcomes between lenalidomide (LEN) and bortezomib (BOR)-based therapies in contrast to other interventions, treatment type was dichotomized to LEN/BOR based vs other. These factors were considered based on evidence of their prognostic value [23, 24] and data availability. Optimal models were selected based on model fit statistics such as Akaike information criteria (AIC), Bayesian information criteria (BIC), R2 and adjusted R2, and mean squared error. All statistical analyses were performed using RStudio version 1.2.5033.
Results
Search results
The SLR identified 91 individual abstracts and manuscripts from 75 original studies consisting of 19 RCTs, 15 of which were phase 3 studies, and 7 single arm studies (Fig 1). Data were extracted from 137 study arms (N = 23,352) with populations ranging from 11 to 1,021 patients per arm. Median PFS was reported in 132 study arms, and reporting of the other endpoints of interest varied, with 124 reporting ORR, 122 CR, 104 very good partial response (VGPR), 31 sCR, and 31 MRD. The other endpoints expressed in terms of months, duration of response (DOR), time to next therapy (TNT), event-free survival (EFS), and time to progression (TTP), were reported in 25 or fewer study arms (Table 1). Based on the available endpoint reporting in the selected studies, ORR, CR, and VGPR had sufficient data for surrogate endpoint validation. VGPR alone does not include patients who reach CR; therefore, ORR is a preferable response-based surrogate endpoint unless a multi-level association could be constructed which factors in the level of response; as such, results for VGPR are not reported. Because of the clinical interest in sCR and MRD, exploratory analyses were performed for these endpoints.
Statistical analysis
The correlation between ORR and median PFS was low (weighted Pearson r = 0.59). The unadjusted linear model demonstrated reasonable fit to the data weighted by sample size (Fig 2A). LEN/BOR-based therapies were generally associated with higher ORR and longer survival than other therapies. According to the unadjusted model, the association was significant (p<0.001), with each percentage point increase in ORR predicting an average median PFS gain of 0.50 months (95% confidence interval CI: 0.36, 0.64). When the model was adjusted for age and treatment type (LEN/BOR vs other), the gain was reduced to 0.35 months (95% CI: 0.21, 0.49).
Scatterplots depict the unadjusted linear model overlayed on the raw data, with point size indicating the relative sample size of each study arm for A) PFS vs ORR and B) PFS vs CR. Dashed lines demonstrate the interpretation of the unadjusted model coefficients. CI, confidence interval; CR, complete response; BOR, bortezomib; LEN, lenalidomide; ORR, overall response rate; PFS, progression-free survival.
Rates of CR were generally low and did not differ substantially between LEN/BOR-based regimens and other therapies. The unadjusted linear model demonstrated reasonable fit to the data weighted by sample size and the correlation between CR and median PFS was low (weighted Pearson r = 0.48) (Fig 2B). Each percentage point increase in CR predicted a median PFS gain of 0.42 months (95% CI: 0.25, 0.58) on average according to the unadjusted model (Table 2). After adjustment for age and ISS risk stage, the associations remained significant but were attenuated to 0.29 months (95% CI: 0.16, 0.41) of median PFS for each point of CR.
Exploratory analyses
Despite limited data, analyses of sCR and MRD were conducted due to current interest in—and clinical relevance of—these endpoints (Fig 3, Table 3). Correlations with median PFS were low for both sCR (weighted Pearson r = 0.68) and MRD (weighted Pearson r = 0.69); however, the associations were statistically significant. Increased sCR and MRD both predict significantly longer median PFS, with unadjusted models predicting an additional 5.25 months and 8.75 months of median PFS per 5% increase in sCR and 25% increase in MRD, respectively. Due to limited sample size, adjusted models could not be constructed.
Scatterplots depict the unadjusted linear model overlayed on the raw data, with point size indicating the relative sample size of each study arm for A) PFS vs sCR (%) and B) PFS vs MRD (%). Dashed lines demonstrate the interpretation of the unadjusted model coefficients. CI, confidence interval; BOR, bortezomib; LEN, lenalidomide; MRD, minimal residual disease; ORR, overall response rate; PFS, progression-free survival; sCR, stringent complete response.
The RWE SLR identified 49 relevant studies reporting on NDMM, representing a total of 10,082 patients. The majority were retrospective (N = 37), and most were single center (N = 30). Substantial heterogeneity with respect to rates of sCR (range: 0 to 46%; N = 25 studies) and MRD (range:10% to 75%; N = 19 studies) was observed in identified studies. Among 20 studies that investigated correlations of OS/PFS with MRD/sCR, 12 reported a significant survival benefit (p<0.05) associated with MRD negativity, and two demonstrated significantly increased survival among patients with sCR.
Discussion
This meta-analysis revealed statistically significant associations, with low correlation, between ORR and CR and median PFS in NDMM, providing evidence to support the use of these outcomes as surrogate endpoints to demonstrate PFS benefit in this population. According to our adjusted model analysis, a 10% ORR or CR increase predicts an incremental median PFS gain of 3.5 months and 2.9 months, respectively. While OS is the gold standard of primary endpoints for demonstrating a survival benefit in oncology trials, PFS is a well-accepted surrogate for OS that can provide more meaningful information on the impact of a first-line treatment [4, 7, 13, 25]. Because patients often survive for several years following initial treatment, median OS is often not reached in the NDMM setting and is subject to the confounding effect of numerous subsequent therapies because patients generally receive many rounds of therapy. Among identified studies in our SLR, a smaller total number reported OS data than PFS data because the median OS was not reached at the time of study publication. Based on the limited results (39 study arms) reporting median OS, our exploratory analysis of reported endpoints and OS showed a low correlation between median OS and CR. This finding may be attributable to reporting bias, wherein studies with high CR were not identified as median OS was not reached.
Recently, both sCR and MRD have emerged as clinically relevant endpoints of increasing interest in MM [18, 19, 26]. Our exploratory analysis of sCR and MRD demonstrated low correlations with statistically significant associations between both endpoints and median PFS, but adjusted models could not be constructed for sCR or MRD due to small sample size. For MRD in particular, the thresholds defined and reported in the identified studies varied from 10−4 (n = 7 studies) to 10−6 (n = 5 studies) and also were derived from differing measurement techniques. These factors, coupled with limited reporting, prevented us from performing additional sensitivity analyses or constructing a multivariable model that adjusted for the level of MRD negativity or measurement technique. This is an area for additional research as such data become available. An examination of RWE supports both the increasing recognition of these outcomes as treatment goals and their association with survival benefit; moreover, the search findings were consistent with the results of our analysis, providing further support for the utility of these endpoints as viable surrogates for PFS in NDMM.
The results of RCT findings and this analysis are aligned with the evidence in other types of cancer. Between 2006 and 2018, the U.S. Food and Drug Administration approved 59 therapies for 85 adult oncology indications on the basis of response rates [27]. Response rates have been associated with improved survival and have demonstrated promise as surrogate markers for survival in other cancers, including acute myeloid leukemia, renal cell carcinoma, and non-small cell lung, breast, and colorectal cancer [28–32]. According to an SLR of meta-analyses assessing response endpoints as surrogates for PFS or OS, there was wide variation [33], underscoring the need for additional studies investigating the relationship between relevant endpoints in specific indications.
While this study presents a comprehensive assessment of response endpoints as surrogates for survival outcomes, it has several limitations. Surrogate endpoint validation in each population was limited by small sample sizes and incomplete reporting of endpoints and patient characteristics. Several important prognostic variables such as Eastern Cooperative Oncology Group performance status and cytogenetic risk were not reported consistently, which limited variable selection for adjusted regression models. Furthermore, certain variables, such as high cytogenetic risk and MRD negativity were defined and measured differently across trials. With respect to high-risk cytogenetics, most studies included 17p-, t4;14, and t14;16; however, others had broader definitions including amp1q21, 13q-, 1p-, 1q+. This may have led to variability that we were unable to adjust for in our analyses due to limited sample size for the various definitions of MRD negativity and high-risk cytogenetics. The possibility of unmeasured confounding remains if factors associated with both surrogate endpoints and survival were unreported or imbalanced between trials and/or study arms. As a result, the conclusions of our study remain limited to the available studies with complete reporting of patient baseline characteristics. While treatment type was considered in the adjusted models, there was insufficient data with respect to chimeric antigen receptor T-cell (CAR T-cell) therapy to account for the possible disease-modifying effect of this potentially transformative treatment [34, 35]. Finally, analyses were conducted using aggregate data, and the observed correlations between intermediate endpoints and PFS may not hold at the individual patient level (ie, ecological fallacy) [36].
We identified low correlation but statistically significant associations between key response endpoints and PFS. According to IQWiG definitions, any R value below 0.7 constitutes a low correlation. The stringency of this threshold is such that the vast majority of identified surrogate endpoint studies do not meet this criteria, including surrogate endpoints that have been accepted by the U.S. Food and Drug Administration for drug licensure or approval [7, 33, 37].
In an SLR of endpoints used in MM RCTs by Mohyuddin et al (2021), only 10 out of 76 (13.2%) phase 3 studies in the NDMM setting employed OS as a primary endpoint; PFS was the most common primary endpoint [38]. Response rate was the most commonly used primary endpoint in identified phase 2 studies [38]. Because of the extended PFS observed in studies of patients with NDMM, a requirement for mature PFS data places a substantial burden on patients who are awaiting access to new therapies. Response endpoints have a long-standing role in allowing for accelerated assessment and approval of promising new therapies in oncology, with the expectation of confirmatory trials that will show evidence of benefit according to survival endpoints. The results of this comprehensive analysis demonstrated a significant association between surrogate endpoints, including ORR and CR, and PFS in NDMM. However, as response rates are already quite high in NDMM, there may be a decreasing margin for improvement remaining which motivates a shift toward more stringent measures of response such as MRD and sCR as surrogate endpoints.
Supporting information
S1 Table. SLR PICOS inclusion/exclusion criteria.
https://doi.org/10.1371/journal.pone.0267979.s001
(DOCX)
Acknowledgments
The authors wish to thank Elizabeth Hubscher, PhD and Claudine Bodin, MS, both of Purple Squirrel Economics, for manuscript writing support and for copyediting and proofreading assistance, respectively.
References
- 1. Kumar SK, Rajkumar V, Kyle RA, van Duin M, Sonneveld P, Mateos M-V, et al. Multiple myeloma. Nat Rev Dis Primer. 2017 Dec;3(1):17046.
- 2.
SEER. Myeloma—Cancer Stat Facts [Internet]. SEER. 2021 [cited 2021 Feb 3]. Available from: https://seer.cancer.gov/statfacts/html/mulmy.html
- 3. Hokland P, Cotter F. Readying the minimal residual disease concept in acute myeloid leukaemia for prime time—the American way. Br J Haematol. 2013 Aug;162(4):429–30.
- 4. Holstein SA, Suman VJ, McCarthy PL. Should Overall Survival Remain an Endpoint for Multiple Myeloma Trials? Curr Hematol Malig Rep. 2019 Feb;14(1):31–8.
- 5. Kumar SK, Callander NS, Hillengass J, Liedtke M, Baljevic M, Campagnaro E, et al. NCCN Guidelines Insights: Multiple Myeloma, Version 7.2021 [Internet]. 2021 Apr. Available from: https://www.nccn.org/professionals/physician_gls/pdf/myeloma.pdf
- 6. Dimopoulos MA, Moreau P, Terpos E, Mateos MV, Zweegman S, Cook G, et al. Multiple myeloma: EHA-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann Oncol. 2021 Mar 1;32(3):309–22.
- 7.
Center for Drug Evaluation and Research. Table of Surrogate Endpoints That Were the Basis of Drug Approval or Licensure. FDA [Internet]. 2021 Mar 31 [cited 2021 May 18]; Available from: https://www.fda.gov/drugs/development-resources/table-surrogate-endpoints-were-basis-drug-approval-or-licensure
- 8. Grigore B, Ciani O, Dams F, Federici C, de Groot S, Möllenkamp M, et al. Surrogate Endpoints in Health Technology Assessment: An International Review of Methodological Guidelines. PharmacoEconomics. 2020 Oct 1;38(10):1055–70.
- 9. Di Leo A, Bleiberg H, Buyse M. Overall survival is not a realistic end point for clinical trials of new drugs in advanced solid tumors: a critical assessment based on recently reported phase III trials in colorectal and breast cancer. J Clin Oncol Off J Am Soc Clin Oncol. 2003 May 15;21(10):2045–7.
- 10. Zhuang SH, Xiu L, Elsayed YA. Overall survival: a gold standard in search of a surrogate: the value of progression-free survival and time to progression as end points of drug efficacy. Cancer J Sudbury Mass. 2009 Oct;15(5):395–400.
- 11. Holstein SA, Jung S-H, Richardson PG, Hofmeister CC, Hurd DD, Hassoun H, et al. Updated analysis of CALGB 100104 (Alliance): a randomised phase III study evaluating lenalidomide vs placebo maintenance after single autologous stem cell transplant for multiple myeloma. Lancet Haematol. 2017 Sep;4(9):e431–42.
- 12. Tacchetti P, Patriarca F, Petrucci MT, Galli M, Pantani L, Dozza L, et al. A Triplet Bortezomib- And Immunomodulator-Based Therapy Before And After Double ASCT Improves Overall Survival Of Newly Diagnosed MM Patients: Final Analysis Of Phase 3 GIMEMA-MMY-3006 Study. EHA [Internet]. 2018 Jun 15 [cited 2021 May 18]; Available from: https://library.ehaweb.org/eha/2018/stockholm/214500/paola.tacchetti.a.triplet.bortezomib-.and.immunomodulatorbased.therapy.before.html?f=listing%3D0%2Abrowseby%3D8%2Asortby%3D1%2Asearch%3Dgimema-mmy230062018
- 13. Cartier S, Zhang B, Rosen VM, Zarotsky V, Bartlett JB, Mukhopadhyay P, et al. Relationship between Treatment Effects on Progression-Free Survival and Overall Survival in Multiple Myeloma: A Systematic Review and Meta-Analysis of Published Clinical Trial Data. Oncol Res Treat. 2015;38(3):88–94.
- 14. Usmani SZ, Schjesvold F, Oriol A, Karlin L, Cavo M, Rifkin RM, et al. Pembrolizumab plus lenalidomide and dexamethasone for patients with treatment-naive multiple myeloma (KEYNOTE-185): a randomised, open-label, phase 3 trial. Lancet Haematol. 2019 Sep;6(9):e448–58.
- 15. Mina R, Oliva S, Boccadoro M. Minimal Residual Disease in Multiple Myeloma: State of the Art and Future Perspectives. J Clin Med [Internet]. 2020 Jul 7 [cited 2021 May 11];9(7). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7408660/
- 16. Martinelli G, Terragna C, Zamagni E, Ronconi S, Tosi P, Lemoli RM, et al. Molecular remission after allogeneic or autologous transplantation of hematopoietic stem cells for multiple myeloma. J Clin Oncol Off J Am Soc Clin Oncol. 2000 Jun;18(11):2273–81.
- 17. Kumar S, Paiva B, Anderson KC, Durie B, Landgren O, Moreau P, et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 8;17(8):e328–46.
- 18. Avet-Loiseau H, Ludwig H, Landgren O, Paiva B, Morris C, Yang H, et al. Minimal Residual Disease Status as a Surrogate Endpoint for Progression-free Survival in Newly Diagnosed Multiple Myeloma Studies: A Meta-analysis. Clin Lymphoma Myeloma Leuk. 2020 Jan;20(1):e30–7. pmid:31780415
- 19. Munshi NC, Avet-Loiseau H, Anderson KC, Neri P, Paiva B, Samur M, et al. A large meta-analysis establishes the role of MRD negativity in long-term survival outcomes in patients with multiple myeloma. Blood Adv. 2020 Dec 8;4(23):5988–99. pmid:33284948
- 20. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar 29;n71. pmid:33782057
- 21.
Borenstein M, editor. Introduction to meta-analysis. Chichester, U.K: John Wiley & Sons; 2009. 421 p.
- 22.
IQWiG. Validity of surrogate endpoints in oncology [Internet]. 2011 Nov [cited 2021 Jun 15]. Report No.: A10-05. Available from: https://www.iqwig.de/download/a10-05_executive_summary_v1-1_surrogate_endpoints_in_oncology.pdf?rev=185859
- 23. Chng WJ, Dispenzieri A, Chim C-S, Fonseca R, Goldschmidt H, Lentzsch S, et al. IMWG consensus on risk stratification in multiple myeloma. Leukemia. 2014 Feb;28(2):269–77. pmid:23974982
- 24. Mikhael JR, Dingli D, Roy V, Reeder CB, Buadi FK, Hayman SR, et al. Management of newly diagnosed symptomatic multiple myeloma: updated Mayo Stratification of Myeloma and Risk-Adapted Therapy (mSMART) consensus guidelines 2013. Mayo Clin Proc. 2013 Apr;88(4):360–76. pmid:23541011
- 25. Broglio KR, Berry DA. Detecting an overall survival benefit that is derived from progression-free survival. J Natl Cancer Inst. 2009 Dec 2;101(23):1642–9. pmid:19903805
- 26. Munshi NC, Avet-Loiseau H, Rawstron AC, Owen RG, Child JA, Thakurta A, et al. Minimal residual disease predicts superior survival in patients with multiple myeloma: a meta-analysis. JAMA Oncol. 2017 Jan 1;3(1):28–35. pmid:27632282
- 27. Chen EY, Raghunathan V, Prasad V. An Overview of Cancer Drugs Approved by the US Food and Drug Administration Based on the Surrogate End Point of Response Rate. JAMA Intern Med. 2019 Jul 1;179(7):915–21. pmid:31135822
- 28. Agarwal SK, Mangal N, Menon RM, Freise KJ, Salem AH. Response Rates as Predictors of Overall Survival: A Meta-Analysis of Acute Myeloid Leukemia Trials. J Cancer. 2017;8(9):1562–7. pmid:28775775
- 29. Tsujino K, Shiraishi J, Tsuji T, Kurata T, Kawaguchi T, Kubo A, et al. Is response rate increment obtained by molecular targeted agents related to survival benefit in the phase III trials of advanced cancer? Ann Oncol. 2010 Aug 1;21(8):1668–74. pmid:20064832
- 30. Petrelli F, Barni S. Surrogate end points and postprogression survival in renal cell carcinoma: an analysis of first-line trials with targeted therapies. Clin Genitourin Cancer. 2013 Dec;11(4):385–9. pmid:24095639
- 31. Tsujino K, Kawaguchi T, Kubo A, Aono N, Nakao K, Koh Y, et al. Response rate is associated with prolonged survival in patients with advanced non-small cell lung cancer treated with gefitinib or erlotinib. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer. 2009 Aug;4(8):994–1001. pmid:19633474
- 32. Blumenthal GM, Zhang L, Zhang H, Kazandjian D, Khozin S, Tang S, et al. Milestone Analyses of Immune Checkpoint Inhibitors, Targeted Therapy, and Conventional Therapy in Metastatic Non–Small Cell Lung Cancer Trials. JAMA Oncol [Internet]. 2017 Aug [cited 2021 May 17];3(8). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5824222/ pmid:28617920
- 33. Cooper K, Tappenden P, Cantrell A, Ennis K. A systematic review of meta-analyses assessing the validity of tumour response endpoints as surrogates for progression-free or overall survival in cancer. Br J Cancer. 2020 Nov;123(11):1686–96. pmid:32913287
- 34. Munshi NC, Anderson LD, Shah N, Madduri D, Berdeja J, Lonial S, et al. Idecabtagene Vicleucel in Relapsed and Refractory Multiple Myeloma. N Engl J Med. 2021 Feb 25;384(8):705–16. pmid:33626253
- 35. Madduri D. CARTITUDE-1: Phase 1b/2 Study of Ciltacabtagene Autoleucel, a B-Cell Maturation Antigen–Directed Chimeric Antigen Receptor T Cell Therapy, in Relapsed/Refractory Multiple Myeloma. In ASH; 2020 [cited 2021 Feb 24]. Available from: https://ash.confex.com/ash/2020/webprogram/Paper136307.html
- 36. Baker WL, Michael White C, Cappelleri JC, Kluger J, Coleman CI, From the Health Outcomes, Policy, and Economics (HOPE) Collaborative Group. Understanding heterogeneity in meta-analysis: the role of meta-regression. Int J Clin Pract. 2009 Oct;63(10):1426–34. pmid:19769699
- 37.
IQWiG. Surrogate Endpoints in Oncology [Internet]. 2011 [cited 2021 May 17]. Available from: https://www.iqwig.de/download/a10-05_executive_summary_v1-1_surrogate_endpoints_in_oncology.pdf?rev=185859
- 38. Mohyuddin GR, Koehn K, Abdallah A-O, Sborov DW, Rajkumar SV, Kumar S, et al. Use of endpoints in multiple myeloma randomized controlled trials over the last 15 years: A systematic review. Am J Hematol. 2021;96(6):690–7. pmid:33735453