Clinical laboratory validation of the MCL35 assay for molecular risk stratification of mantle cell lymphoma

Abstract

Mantle cell lymphoma (MCL) is a clinically heterogeneous B cell malignancy for which a variety of prognostic factors have been proposed. Previously, a digital gene expression profiling “proliferation signature” capable of risk stratifying MCL was identified and subsequently developed into a multi-analyte prognostic assay, known as the “MCL35” assay. In this study, we sought to explore the performance characteristics of the MCL35 assay in a clinical laboratory and compare results with the Ki67 proliferation marker. The results describe the clinical validation of the MCL35 assay for molecular risk stratification of MCL including accuracy, sensitivity, specificity, use in acid-decalcified bone marrow core biopsies, fixatives, lower limit of RNA input, quality metrics, and other laboratory parameters. The resulting data indicate that this is a robust technique with outstanding reproducibility. Overall, the data support the concept of molecular signatures, as assessed with digital gene expression profiling, for improved standardization and reproducibility for proliferation assessment in MCL.

Evaluation of clinical parameters and biomarkers in older, untreated mantle cell lymphoma patients receiving bendamustine–rituximab

“…With the prior technical validation of the MCL35 in a clinically accredited laboratory, including an interlab concordance of 100% and meeting the REMARK criteria for biomarkers, our results demonstrate the confidence with which the MCL35 can identify the high-risk category. 9,10,36 Digital gene expression profiling is in use for other haematological tumours. 9,[37][38][39][40][41][42] In this study, both the MCL35 and LMCL16 were 100% successful in all attempted cases.…”

“…9 The MCL35 assay was subsequently validated for accuracy and reproducibility in a CAP-accredited, CLIA-certified molecular diagnostics laboratory. 10 When applied to patients <65 years old receiving intensive chemotherapy on the Nordic MCL2 and MCL3 studies, 11,12 the MCL35 assay and MIPI together successfully risk-stratified patients and identified a 'dismal' prognosis group. 13,14 Several additional variables are used to risk-stratify MCL patients, including cytologic variation.…”

“…Because of deficiencies in the single‐parameter Ki67 IHC test, the Lymphoma/Leukemia Molecular Profiling Project (LLMPP) developed an assay to quantify a proliferation signature with robust performance in formalin‐fixed, paraffin‐embedded tissues (FFPET) known as the ‘MCL35’ 9 . The MCL35 assay was subsequently validated for accuracy and reproducibility in a CAP‐accredited, CLIA‐certified molecular diagnostics laboratory 10 . When applied to patients <65 years old receiving intensive chemotherapy on the Nordic MCL2 and MCL3 studies, 11,12 the MCL35 assay and MIPI together successfully risk‐stratified patients and identified a ‘dismal’ prognosis group 13,14 …”

Artificial Intelligence Analysis of Gene Expression Predicted the Overall Survival of Mantle Cell Lymphoma and a Large Pan-Cancer Series

“…A machine learning and neural network modeling was performed using the highlighted genes of both Methods 1 (19 genes) and Methods 2 (15) with the previously identified prognostic genes of MCL of the LLMPP, the MCL35 signature [50,[67][68][69]. All the available artificial intelligence methods were tested, and high overall accuracy for predicting was found for logistic regression (100%), Bayesian network (92%), discriminant analysis (86%), CHAID (85%), C&R tree (85%), and SVM (81%) (Table 4, Figure 13).…”

Towards precision medicine in lymphoid malignancies

“…In MCL, proliferation signatures have long been described [93] to have an important prognostic impact and can now be investigated in FFPE material [94][95][96][97]. TP53 is also an adverse prognostic factor in MCL [98,99], as well as in DLBCL [100][101][102][103] and LPL/WM [104].…”