Blood-based microRNA signature may predict survival in immunotherapy-treated NSCLC

 March 31, 2022

A blood-based signature of five microRNAs demonstrated the ability to predict OS of patients with lung cancer receiving immunotherapy and stratify them for treatment, according to a study published in npjPrecision Oncology

Researchers reported that because the signature appears to work from a single timepoint before treatment, it differs from classical circulating tumor DNA-based assays, where blood at baseline is typically dynamically compared with follow-up samples.

Quote from Bruno R. Steinkraus, PhD.

“Importantly, the signature does not require any prior tumor tissue analysis and could be used in an ‘off-the-shelf’ manner,” Bruno R. Steinkraus, PhD, chief scientific officer at Hummingbird Diagnostics, told Healio. “We believe our test could form a practical decision aid in identifying the right patient population to treat with immunotherapy.”

Background and methodology

Although immunotherapy has been highly effective in some late-stage cancers, many patients do not benefit from the treatment, and others either fail to respond or experience immune-related adverse events, according to researchers. “There is a pressing need for more accurate biomarkers for immunotherapy response prediction,” they wrote. 

“We believe that the currently used tumor-intrinsic characteristics (such as PD-L1 expression or tumor mutational burden) are only one part of the equation predicting response to immunotherapy,” Steinkraus said. “Since systemic immunity is also required for effective checkpoint inhibition and the immune system is the very effector mechanism these therapies are trying to unleash, we set out to analyze this peripheral immune axis.”

The researchers decided to analyze blood-borne microRNAs because they are molecular orchestrators of gene expression and therefore act as excellent surrogate markers for the constellation and activity of the immune system, according to Steinkraus. 

“The fact that they are nucleic acid-based means they can be amplified by sensitive techniques such as next-generation sequencing and quantitative polymerase chain reaction and analyzed in a massively parallel manner,” Steinkraus said. 

The analysis included 334 patients with stage IV non-small cell lung cancer separated into three cohorts based on time of recruitment and treatment. The training cohort (n = 96; median age, 68.2 years; range, 38.9-86.7; 62.5% men) and independent validation cohort (n = 99; median age, 66.4 years; range, 33.5-87; 64.6% men) consisted of patients who received pembrolizumab (Keytruda, Merck) or nivolumab (Opdivo, Bristol Myers Squibb), whereas patients in the control cohort (n = 139; median age, 64.9 years; range, 37.6-84.8; 66.9% men) received platinum doublet chemotherapy and pembrolizumab.

Key findings 

Researchers defined a five microRNA risk score predictive of OS after immunotherapy in both the training and independent validation cohorts (HR = 2.4, 95% CI, 1.37–4.19).

Steinkraus noted their signature microRNAs largely originated from circulating myeloid cells.

“Oncology has mainly been concerned with lymphoid cells, particularly T cells. However, there is a growing body of evidence that myeloid cells are important as first responders and can actively configure the tumor microenvironment and, thus, potentially influence the effectivity of the antitumor T-cell response,” he said.

Through microRNA target prediction, researchers made a direct mechanistic connection to the PD-L1 signaling pathway and PD-L1. They concluded that the microRNA risk score “offers a potential blood-based companion diagnostic for immunotherapy that outperforms tissue-based PD-L1 staining.”


Steinkraus and colleagues hope to investigate whether the signature works in other cancers where immunotherapy is utilized, including bladder cancer.

“Hummingbird is committed to take this proof-of-concept work to the bedside,” he told Healio. “We have identified the treatment decision of immunotherapy alone vs. immunochemotherapy combination therapy in PD-L1-high patients as our first clinical use case and are currently preparing further trials.”

For more information:

Bruno R. Steinkraus, PhD, can be reached at Hummingbird Diagnostics GmbH, Im Neuenheimer Feld 583, 69120 Heidelberg, Germany; email: