28.4. Prognostic value of sFLCs at diagnosis

Chapter 28

In addition to assessing the degree of organ involvement [11][611], evaluation of sFLCs at baseline provides important prognostic information in AL amyloidosis, and is recommended in IMWG guidelines (Section 25.3.1.) [167]. In a study of 730 patients, median overall survival was shorter among those with an abnormal sFLC ratio at baseline (16.2 months, n=644) than in those with a normal ratio (63.6 months, n = 86) (Figure 28.8A ) [192]. When the analysis was repeated grouping patients according to sFLC burden (defined as the difference between involved and uninvolved FLC [dFLC]) above or below the median value (196 mg/L), the overall survival for patients with high dFLC was 10.9 months compared with 37.1 months for those with low dFLC (p<0.001) (Figure 28.8B ) [192]. High baseline FLC levels have also been shown to be associated with poor outcome in AL amyloidosis patients undergoing stem cell transplant [11][616], regardless of treatment type [618].

Kumar et al. [192] noted that AL amyloidosis patients had a worse survival if there was no intact monoclonal immunoglobulin detectable (12.6 months vs 29.3 months, p=0.02). The patients without monoclonal heavy chain had a higher dFLC value (255 vs 153 mg/L, p<0.001) but on multivariate analysis both dFLC and the presence/absence of heavy chain were independently prognostic for survival. Kumar et al. [192] also reported that AL amyloidosis patients with high sFLC burden (dFLC >196 mg/L), had more frequent and severe cardiac involvement, with higher levels of cardiac biomarkers troponin T (cTnT) and pro-B-type natriuretic peptide (NT-ProBNP) . However, in a multivariate analysis that included a variety of markers, baseline dFLC remained an independent predictor of survival. In a subsequent publication from the same centre [11], a revised staging system for AL amyloidosis was proposed incorporating dFLC alongside the cardiac biomarkers cTnT and NT-ProBNP. Patients were assigned a score of 1 for each of the following: 1) dFLC ≥180 mg/L; 2) cTnT ≥0.025 ng/mL or, 3) NT-ProBNP ≥1800pg/mL, to give stages I to IV (for 0 - 3 points, respectively). The staging system was derived from 810 patients with newly diagnosed AL amyloidosis and tested on two further populations (n=303 and n=103). The authors concluded that this revised Mayo staging system (based on sFLC measurements and cardiac biomarkers) allowed better discrimination of patient outcomes in AL amyloidosis than a staging system based on cardiac biomarkers alone, and should be incorporated into future clinical trials [617].

Three independent studies have characterised the subgroup of AL amyloidosis patients who do not have measurable disease by sFLCs at diagnosis (dFLC <50 mg/L) [947][949][948]. This group accounts for 15 - 19% of all newly diagnosed patients, and represents a distinct clinical entity. Milani et al. [947] reported that such patients are characterised by a lower incidence of heart involvement (43% vs. 83%, p<0.001), higher incidence of renal involvement (77% vs. 63%, p<0.001) that tended to be more severe, with a more favourable outcome compared to with patients with evaluable FLCs (median 117 vs. 21 months, p<0.001). Similar findings were observed by Dittrich et al. [949]. This study also reported lower bone marrow plasma cell counts (6% vs. 10%, p<0.001), smaller M-spikes (7 g/L vs. 9 g/L (p<0.001) and lower concentrations of monoclonal heavy chain (7 g/L vs. 10 g/L, p=0.003) among patients with a dFLC <50 mg/L. It has been suggested that such patients are underrepresented in clinical studies, because their haematologic response cannot be reliably measured, which may adversely influence results [618]. However, Milani et al. [947] have demonstrated a survival benefit for patients with a low dFLC burden who subsequently achieve a complete response, and proposed that these patients can be included in clinical trials with appropriate stratification and be evaluated for complete response.

Wassef et al. [608] used the above staging system when comparing sFLC results produced by Freelite with another manufacturer's monoclonal antibody-based FLC assays. The quantitative results and stage allocation differed for the two immunoassays, indicating that the staging system cannot be extended to different assays without separate validation (Chapter 8).