Many patients with EGFR-mutated non-small cell lung cancer, or NSCLC, face a frustrating prognosis. While initial treatment with EGFR-targeted tyrosine kinase inhibitors (TKIs) works well, “persister” cells that are left behind will almost certainly cause the cancer to return. The disease becomes much harder to fight when it comes back, as the new tumors are highly resistant not only to TKIs but other drugs as well.
But scientists at the University of Texas MD Anderson Center have found a way to wipe out leftover EGFR-mutated NSCLC cells before they develop complete resistance. Most NSCLC cases are diagnosed at an advanced stage, and EGFR mutations are present in about 10% to 15% of them, according to MD Anderson.
In a new study published Feb. 13 in Cancer Cell, they described how they used cell therapies and antibody-drug conjugates (ADCs) targeted at the protein CD70 to eliminate residual EGFR-mutated NSCLC cells in cell lines and animal models.
“The finding that CD70 is present on the most aggressive forms of EGFR TKI resistance as well as on the residual cells after initial TKI treatment raises the possibility that we could combine an EGFR inhibitor with CD70-targeting agents, which would eliminate the residual persister cells,” John Heymach, M.D., Ph.D., corresponding author and chair of Thoracic/Head and Neck Medical Oncology at MD Anderson, told Fierce Biotech Research in an email. “We believe this combined approach could dramatically delay or prevent the emergence of drug resistance.”
First-line TKI treatments such as AstraZeneca’s Tagrisso (osimertinib) work well in killing off the initial tumors. Over time, the cells that are inevitably left behind can resume expansive growth, sometimes through a process called epithelial to mesenchymal transition, which ultimately results in tumor formation. Virtually all patients whose EGFR-mutated NSCLC is first diagnosed and treated in stage 4 can expect their cancer to come back.
The researchers wondered whether there was a way to intervene on the recurrence process by identifying and targeting unique proteins on the surface of residual cells. To find out, they cultured EGFR-mutated NSCLC cell lines in the presence of EGFR inhibitors until the cells developed resistance to them. They then conducted RNA sequencing, a technique that reveals what genes are expressed in a cell or biological sample, to see which cell surface proteins were being transcribed.
Their search turned up two possible targets: the genes for the proteins EMP3 and CD70. However, EMP3 was also present on healthy tissues in the lung, kidneys and GI tract. CD70 expression, on the other hand, was limited to activated lymphocytes and the resistant cells. This made it a better candidate for further study, as targeting CD70-expressing cells would come with a lower risk of side effects.
Additional cell studies verified that CD70 was present on treatment-resistant cells from NSCLC patients, and that a high amount of CD70 expression was associated with worse overall survival rates. But while the scientists found that epithelial to mesenchymal transition did indeed upregulate CD70 levels—and that CD70 itself caused tumors to grow by activating key signaling pathways—the scientists also found that it was present in cells that seemed to have begun epithelial to mesenchymal transition but had not yet developed full resistance to TKIs.
“We were surprised to see that CD70 was upregulated rapidly after TKI treatment in the drug-tolerant persister population,” Monique Nilsson, Ph.D., who led the study, said. “This raises the possibility that CD70 could be targeted early in NSCLC patients with EGFR mutant disease rather than waiting for aggressive, resistant disease to fully emerge.”
The researchers then tested the efficacy of different CD70-based therapeutic strategies against both the resistant and not-yet-resistant persister cells in a petri dish. They found that an their own formulation of the ADC CD70-antibody cusatuzumab conjugated to the chemotherapy agent toxin monomethyl auristatin E was effective in most cell lines, while CD70-targeted CAR T cells and CAR natural killer cells (CAR NKs) were effective against all of them.
Finally, the researchers tested their findings in animal models. They injected different groups of mice with the EGFR TKI resistant cells, then after they developed tumors treated them with either custauzumab-MMAE or CD70 CAR NK cells. Both treatments caused the tumors to regress, and they hadn’t rebounded by the end of the 60 to 80 day study period.
The researchers believe the treatments they tested warrant further study, especially the CAR-based therapies. While all the drug resistant cells were vulnerable to CD70 CAR-NKs and CAR-Ts, a few cell lines weren’t sensitive to custauzumab-MMAE. The scientists’ data suggested that this was because they were resistant to the chemotherapy portion of the drug, a problem that could arise with other types of ADCs too. On top of that, some payloads used in ADCs have the greatest amount of activity on dividing cells, so for targeting these early persister cells, which are senescent, it will be important to select an ADC with the optimal payload.
Clinical trials on CD70-based CAR therapies and ADCs are already underway for kidney cancer, pancreatic cancer and others, and the scientists hope to add NSCLC to the list. They also believe that their findings about EGFR inhibitor resistance could apply to other types of cancer where persister cells eventually undergo epithelial to mesenchymal transition along with other treatment resistance mechanisms.
Editor's note: This article has been updated to attribute a quote to Heymach and to clarify the experiments and results.