A genome-forward approach to tackling drug-resistant cancers

Overall, the PDX approach promises to speed the development of new drugs along with doctors' ability to make more precise choices about how those drugs are used to treat patients, the researchers say.

"The development of precision pharmacology is clearly the current focus in PDX research," said Matthew Ellis of Washington University in St Louis. "Human testing is hugely expensive, and often the response rates for the patients on experimental drugs are low because the biology of each patient is not well defined. Panels of clinically and genomically annotated PDX can therefore be very valuable for studying drug action and developing predictive biomarkers. Extensive pre- and post-drug sampling can be conducted to study drug effects and drug resistance in a way that would be impossible in the clinical setting."

In the new study, Ellis and his team transplanted drug-resistant human breast cancers into mice and then made very detailed comparisons of those transplanted tumors versus the originals.

The researchers' deep whole-genome analyses showed a high degree of genomic fidelity. In other words, the complex human tumor tissues in the mice looked very much like those in the people they originally came from. While some new mutations did arise after transplantation, those genetic changes rarely had functional significance.

The researchers were surprised to discover that the original and PDX cancers were similar at the cellular level as well. Cancer cells carrying mutations that were relatively rare in the patient were also maintained at lower frequencies in the mice. Likewise, more dominant clones in the original tumor tended to stay dominant in the mice. This suggests that the frequency of genetically distinct tumor cells is in an equilibrium that survives transplantation into mice for reasons that aren't yet clear.

An analysis of multiple estrogen receptor-positive PDX from patients with endocrine therapy-resistant disease shows just how this approach can yield tumor-specific explanations for therapy resistance. Resistant tumors were associated with different kinds of alterations to the estrogen receptor gene ESR1, the researchers found, producing different responses to endocrine therapy.

"The prevalence of ESR1 mutations and gene arrangements in the luminal PDX was a deep surprise to me as I thought these events were rare," Ellis said. "There had been very sporadic reports of ESR1 point mutations in clinical samples over the years, but to find them at high prevalence in the PDX and therefore in a setting where the link to endocrine therapy resistance can be directly studied was, for me, a critical breakthrough in our understanding of this critical problem."