Page 180 - 2019_07 resto del Mondo-web

T. Dittrich et al.
did not have a favorable prognosis compared to that of patients assigned to the highest stage without these con- ditions. Indeed, patients with eGFR<50 classified as high- est risk by the MAYO2004 and MAYO2012 staging sys- tems had an even worse prognosis than that of patients with eGFR ≥50 mL/min/1.73 m2, while there was no such difference in the MAYO3b system. This suggests that the higher stages due to the elevation of cardiac biomarkers are correctly assigned and provides evidence that the MAYO3b staging system is superior to the MAYO2004 and MAYO2012 systems in patients with impaired kidney function.
(II) Further evidence for a superiority of the MAYO3b system comes from the evaluation of the additional prog- nostic value of eGFR<50 and AF in the context of the stag- ing systems. In multivariable analyses including age, dFLC and any of the staging systems, eGFR<50 and AF each retained their prognostic significance with the MAYO2004 and MAYO2012 systems, while eGFR<50 fell out with MAYO3b.
(III) We further evaluated the performance of the staging systems by synopsis of stratified Kaplan-Meier estimates (Figure 2) as well as predictive and discriminative accuracy using prediction errors and concordance indices (Figure 3). We confirmed the high prognostic value of each of the staging systems in our entire cohort. However, the MAYO3b and MAYO2012 systems showed a better dis- criminative accuracy than the MAYO2004 system, and the MAYO3b system was superior in detecting patients with the best and worst prognoses. In the subgroup with low eGFR, stages I and II within any of the staging systems could not discriminate patients with different survival. This did not, however, affect the overall predictive and discriminative accuracy of all staging systems, which again confirms that patients with eGFR<50 were correctly assigned to higher stages. In the AF subgroup, the systems were less precise, potentially also due to low numbers of patients assigned to the low-risk stages. The MAYO2012 system, in particular, lost discriminative accuracy, making the MAYO3b system superior in this subset. This might be because patients with very high risk of early death are well discriminated by stage IIIb, while in the MAYO2012 system the median survival of patients in the two highest stages is not significantly different. Hence, very high levels of NT-proBNP are prognostic in AL patients irrespective of eGFR or AF status. Herewith, we confirm recent findings made in other cohorts of patients with heart failure.26,34
Overall, our results suggest that the MAYO3b staging system should currently be preferred over the other wide-
ly used systems for the following reasons: (i) it showed the best performance and highest discriminative accuracy in the entire cohort; (ii) it was most applicable in the high- risk subsets of patients with eGFR<50 or AF; and (iii) it only requires two biomarkers compared to the three bio- markers required for the MAYO2012 score, which is an advantage from the perspective of model optimization.
A recently proposed staging system for cardiac transthyretin amyloidosis is based on NT-proBNP and eGFR.35 Our results suggest, however, that the prognostic potential of reduced eGFR is already included in a staging system with multiple NT-proBNP thresholds such as MAYO3b. Interestingly, the superiority of the modified MAYO staging system with two NT-proBNP thresholds was also reported by the Mayo group itself at the annual International Symposium on Amyloidosis in March 2018 following calculation of the net reclassification improve- ment for pairs of prognostic models.36
On the other hand, our multivariable analysis suggests that the significant prognostic potential of atrial arrhyth- mia and dFLC is not entirely contained in the MAYO3b system. This is supported by the fact that the MAYO2012 system was able to identify patients with different out- comes from each stage of the MAYO3b system. In addi- tion to AF and plasma cell-related factors, such as dFLC,21 the inclusion of other promising and widely available bio- markers such as serum uric acid22 might further improve the MAYO3b system. In contrast to novel cardiac bio- markers that have recently been suggested for inclusion into the AL amyloidosis staging system,19 these parame- ters are readily available in every newly diagnosed AL patient worldwide. Finally, a potential modification of the MAYO3b staging system and validation of a novel staging system might be best performed with a multicenter approach, e.g. on behalf of the International Society of Amyloidosis, to achieve greatest validity and facilitate broad acceptance.
To our knowledge, this is the first report that extensive- ly describes kidney impairment and atrial arrhythmia as independent prognostic factors in AL amyloidosis. Most importantly, the detailed performance analysis of the AL staging systems, including high-risk subgroups, will enable clinicians and investigators to make a confident choice within the broad range of currently available stag- ing systems.
Acknowledgments
The authors thank Rita Ziehl, medical data manager, for her assistance with the collection of clinical data.
References
1. Maurer MS, Elliott P, Comenzo R, Semigran M, Rapezzi C. Addressing common ques- tions encountered in the diagnosis and man- agement of cardiac amyloidosis. Circulation. 2017;135(14):1357–1377.
2. Lisenko K, Schönland SO, Jauch A, et al. Flow cytometry-based characterization of underlying clonal B and plasma cells in patients with light chain amyloidosis. Cancer Med. 2016;5(7):1464–1472.
3. Comenzo RL, Reece D, Palladini G, et al. Consensus guidelines for the conduct and reporting of clinical trials in systemic light-
chain amyloidosis. Leukemia. 2012;26(11):
2317–2325.
4. Dittrich T, Bochtler T, Kimmich C, et al. AL
amyloidosis patients with low amyloido- genic free light chain levels at first diagnosis have an excellent prognosis. Blood. 2017;130(5):632-642.
5. Abraham RS, Katzmann JA, Clark RJ, Bradwell AR, Kyle RA, Gertz MA. Quantitative analysis of serum free light chains. A new marker for the diagnostic evaluation of primary systemic amyloidosis. Am J Clin Pathol. 2003;119(2):274–278.
6. Milani P, Basset M, Russo F, Foli A, Merlini G, Palladini G. Patients with light-chain
amyloidosis and low free light-chain burden have distinct clinical features and outcome. Blood. 2017;130(5):625–631.
7. Kimmich C, Schönland S, Kräker S, et al. Amyloid in bone marrow smears in sys- temic light-chain amyloidosis. Amyloid. 2017;24(1):52–59.
8. Sachchithanantham S, Hutt DF, Quigley A- M, Hawkins P, Wechalekar AD. Role of 99mTc-DPD scintigraphy in imaging extra- cardiac light chain (AL) amyloidosis. Br J Haematol. 2018;183(3):506-509.
9. Gertz MA, Kyle RA. Hepatic amyloidosis: clinical appraisal in 77 patients. Hepatology. 1997;25(1):118–121.
1458
haematologica | 2019; 104(7)