HGF Measurement in AL Amyloidosis.

Hepatocyte Growth Factor (HGF) is a pro-angiogenic cytokine and a mitogenic, motogenic and morphogenic factor involved in tumor growth. Previous studies have shown that HGF is secreted by plasma cells in multiple myeloma and that HGF serum levels are higher in patients with multiple myeloma and correlate with disease activity. A previous study reported that serum HGF levels were significantly higher in patients with AL amyloidosis compared to patients with multiple myeloma (Iwasaki et al. Br J Haematol. 2002;116:796-802). A preliminary study of 18 AA and AL amyloidosis patients (Shikano et al, Intern Med. 2000;39:715-9) suggested that measurement of HGF might be useful for the diagnosis of amyloidosis. To determine whether HGF may be used as a relevant diagnosis marker and prognosis factor in AL amyloidosis, we have measured HGF serum levels in patients with AL amyloidosis and patients with plasma cell dyscrasia without amyloidosis.

Two groups of patients were included; patients with biopsy proven AL amyloidosis and patients with plasma cell dyscrasia (MGUS, multiple myeloma, POEMS) without amyloidosis as controls. Levels of HGF were measured by ELISA at diagnosis in the two groups, before any treatment (Quantikine® R&D Systems). Clinical features were recorded for AL patients. A Receiver Operating Characteristic curve (ROC) analysis was performed to assess the diagnostic accuracy of HGF for identification of amyloidosis cases among patients with monoclonal gammopathy. The area under the ROC curve (AUC) which can be interpreted as the probability that a randomly chosen amyloidosis patient has a test result greater than that of a randomly chosen non-amyloidosis patient, was calculated with its 95% confidence interval (95%CI). The ROC curve was also used to determine the best threshold for HGF. Using this threshold, sensitivity and specificity were calculated. Survival analyses were performed for patients suffering from AL amyloidosis. Baseline time was time from first HGF assessment to death or censoring date. Univariate analysis were done using Kaplan Meier and Cox proportional hazard models.

Sixty-nine AL amyloidosis patients diagnosed between 2004 and 2008 and 76 controls (56 patients with MGUS, 17 with multiple myeloma, three with POEMS) were included. The median age was 61 (32-90) for AL patients and 60 (39-86) for controls. Median creatinine levels were respectively 86μmol/l (39-500) and 79μmol/l (44-317); 57 AL patients (82.6%) had renal involvement and 40 had (57.9%) cardiac disease. Monoclonal protein isotype was lambda in 69.6% of AL patients and kappa in 30.4%. HGF serum levels were significantly higher in patients with AL amyloidosis: 11.2ng/ml (0.5-200.4) compared with controls: 1.5ng/ml (0.8-8.2), p<0.0001 (healthy controls 0.9 ng/ml). HGF levels at diagnosis seemed to be discriminant with area under the ROC curve at 0.896 IC95% [0.834-0.94] p=0.0001. The threshold value of 2.4ng/ml conferred the best sensitivity : 82.6% IC95% [71.6-90.7] and specificity : 89.5% IC95% [80.3-95.3] for the diagnosis of AL amyloidosis. Patients were treated mainly by conventional chemotherapy (M-Dex), 65 % of AL patients were alive after a median follow up of 18 months. Univariate analysis showed a relative risk of mortality of 1.70 in AL patients with HGF levels upper than 11ng/ml, compared to those with HGF levels under 11 ng/l who showed a trend for better survival (p=0.22).

This study confirms that HGF levels are elevated in patients with AL amyloidosis, significantly higher than in patients with other plasma cell disorders. A threshold value of 2.4ng/ml confers a good sensitivity (80%) and specificity (90%) to suggest AL amyloidosis. HGF measurement may be used in patients with plasma cell dyscrasia to determine which patient should be considered for a biopsy. We found a trend towards reduced survival in patients with the highest levels of HGF. This, and the usefulness of HGF measurement in predicting clinical responses should be confirmed on larger studies.

No relevant conflicts of interest to declare.