NEW YORK (GenomeWeb News) – A gene expression signature associated with activation of the tumor necrosis factor CD40 pathway can help predict response to a new B cell cancer treatment targeting the pathway, according to an online study in Science Translational Medicine today.
A Genentech-led team used a combination of targeted sequencing and microarray analyses in non-Hodgkin's lymphoma cell lines to find mutation and gene expression patterns linked to resistance or sensitivity to dacetuzumab, a CD40-stimulating antibody being investigated for the treatment of diffuse large B cell lymphomas and other B cell cancers.
Among the expression patterns associated with resistance was activation of a pathway containing CD40 itself, they found. That, in turn, prompted the researchers to develop an assay based on a CD40 activation signature that they subsequently used to predict dacetuzumab treatment response in cell lines, mouse xenograft models, and clinical samples.
"These data show that CD40 pathway activation status may be useful in predicting the anti-tumor activity of CD40-stimulating therapeutic drugs," senior author David Dornan, a molecular diagnostics and cancer cell biology researcher at Genentech, and co-authors wrote.
B cells are a type of white blood cell that contribute to adaptive immune system function by producing and tweaking antibodies. Receptors on the surface of these cells, including CD40 and other proteins in the tumor necrosis factor receptor super-family, contribute to T cell immune function and transmit signals to the B cells that help keep them alive, researchers explained.
On the other hand, activation of the same pathway seems to spell doom for certain cancerous B cells, they added, making the CD40 pathway activation an attractive strategy for treating some B cell cancers.
"Stimulation of the CD40 pathway promotes apoptosis of certain B cell lines in vitro and in xenograft tumor models," they noted. "In contrast, CD40 stimulation can cause proliferation of non-Hodgkin's lymphoma cells and promote lymphomagenesis and the transformation of primary B cells."
In an effort to find ways to distinguish between cancers that are sensitive or resistant to CD40-stimulating treatment, the team trolled through a panel of non-Hodgkin's lymphoma cell lines, classifying them as sensitive or resistant to a CD40-stimulating monoclonal antibody known as dacetuzumab. Most of the lines had been generated from diffuse large B cell lymphomas, they noted.
The team then characterized somatic mutations in the lines by sequencing 21 oncogene and tumor suppressor genes — findings they reported in another study — and used flow cytometry to look for evidence of DNA damage in the cells. Gene expression patterns, meanwhile, were evaluated using the Affymetrix HGU133Plus_2.0 and Affymetrix Gene 1.0ST arrays.
Overall, researches found, cell lines harboring mutations in the tumor suppressor gene p53 were more likely to respond to the CD40 stimulation treatment, as were cell lines with loads of DNA damage or unusually high proliferation rates. Elevated levels of the transcriptional repressor proto-oncogene BCL6 also coincided with treatment sensitivity.
In contrast, cell lines exhibiting expression patterns consistent with CD40 pathway activation prior to treatment tended to be resistant.
Based on these findings, the team came up with a quantitative reverse transcription-PCR assay for trying to predict sensitivity to dacetuzumab. They first tested this qRT-PCR assay — which assesses the expression of 15 genes, including BCL6 and genes in the CD40 pathway — by using it to predict dacetuzumab sensitivity or resistance in non-Hodgkin's lymphoma cell lines.
In a training set that included 30 cell lines, the signature accurately predicted treatment susceptibility for 23 cell lines, the researchers reported. In a dozen more non-Hodgkin's lymphoma lines, meanwhile, the signature distinguished between sensitive and resistant lines 75 percent of the time. Similarly, the signature accurately predicted treatment outcomes for three of four xenograft mice whose tumors were profiled prior to implantation.
The team further tweaked their assay using formalin-fixed and paraffin-embedded cell lines before using it to assess 39 clinical samples from diffuse large B cell lymphoma patients treated with dacetuzumab during phase 1 and 2 trials of the therapy.
In these samples, the response signature accurately predict treatment response — specifically, tumor shrinkage — for 80 percent of the cases. Patients classified as having treatment sensitive tumors based on the expression signature also had significantly longer progression-free survival times following dacetuzumab treatment than those with tumors classified as treatment resistant.
"[O]ur findings suggest that anti-tumor responses to CD40 stimulation in [non-Hodgkin's lymphoma] depend on the pre-existing status of the CD40 pathway," the researchers concluded, noting that "[t]he hypothesis that anti-tumor responses of agonist CD40 therapeutics an be predicted by CD40 pathway signatures will need to be further validated in upcoming clinical trials."