Treatment of systemic AL amyloidosis (AL) remains difficult, especially in older and sicker patients in whom dose intensive therapies are often associated with unacceptable morbidity and mortality. Many such patients continue to be treated with oral melphalan ± prednisone (MP) despite early trials having shown only very modest clinical efficacy, presumably encouraged by its perceived low toxicity. The recent advent of the sensitive nephelometric serum free light chain (sFLC) assay has for the first time enabled the typically subtle underlying clonal disease to be monitored in an effective quantitative manner in the majority of patients with AL. We report here sFLC responses and clinical outcome of patients with AL who received MP first-line and underwent serial evaluations at the UK National Amyloidosis Centre. The 90 patients comprised 46 males and 44 females with a median age of 68yrs (range 43–83). Median number of organs involved was 3 (1–4), including kidneys in 72%, heart in 56%, and liver in 29%. 16 (17%) had ≥ NYHA class III heart failure. Median ECOG performance status was 1. Median follow-up was 2.3 yrs (0.3–14). 60 patients received oral melphalan with prednisone, and 30 received single agent melphalan. Patients received a median of 6 cycles of treatment (range 1– 26), and the sFLC assay was scheduled following each cycle after availability of the assay and retrospectively on stored sera for earlier patients. Haematological response data using sFLC assay were evaluable in 54 (60%). Responses were defined as complete response (CR) - sustained normalisation of sFLC ratio, partial response (PR) - sustained ≥50% reduction in pre-treatment clonal isotype. There was a haematological response in 22 (40%) of evaluable patients. 4 (7%) had a complete response, 18 (33%) had a partial response and 32 (59%) did not respond. 42% of patients treated with single agent melphalan responded compared with 39% of those treated with melphalan and prednisone (p=0.8). Responders received a median 6 cycles of treatment, and complete responders received a median of 14 cycles. Non-responders received a median of 5 cycles of treatment. The median time to commencing further chemotherapy was 5 months. The median overall survival (OS) was 5.8yrs, but most patients received further salvage treatments and the influence of MP treatment on OS could not be ascertained. Toxicities during MP were seen in 13 (14%) cases, including myelodysplasia in 2 patients. There were no treatment related deaths. In conclusion, use of the sFLC assay confirms that response of the underlying clonal plasma cell disease to standard oral melphalan and prednisone is poor in AL amyloidosis, and that response is usually very delayed even among patients who respond completely. Encouragingly however, toxicity was low among this relatively old and sick cohort of patients. These findings support frequent sFLC measurements in AL patients receiving MP to enable rational treatment decisions to be made at the earliest opportunity.