“Precision Medicine” is the latest buzzword, and its goal is to individualize care based on the biology of each patient. For cancer therapy, this means taking into consideration myriad molecular and cellular features; namely, genomics, RNAseq, mRNA microarray, microRNA, proteomics, metabolomics, and others. Because each cancer appears to be unique, there might be distinct points of vulnerability that could be exploited for treatment. The ideal would be to not undertreat patients by administering drugs to which their cancer isn’t sensitive, and to not over-treat, causing excessive toxicity and the draining of unnecessary resources. The goal should be nontoxic “magic bullets” that have specific targets and are administered easily (preferably orally), with rapid action and resolution of the tumor, and ultimately, cure of the disease. This has happened previously in the field of leukemia — with tyrosine kinase inhibitors (TKIs) for chronic myelogenous leukemia (CML). With several years having elapsed since the U.S. Food and Drug Administration approval of imatinib and related TKIs, it seems that we are not close to identifying comparable drugs despite ongoing efforts to sequence DNA and RNA, analyze proteins, and test tumor cells against drugs ex vivo. Additionally, all the ex vivo approaches to chemotherapy drug testing have many drawbacks, including lack of the appropriate tumor microenvironment and lack of the relevant organs that metabolize specific drugs.
From this quagmire now emerge ingenious devices that fulfill the promise of how clever engineering will transform medicine. Two concurrent publications in Science Translational Medicine describe the parallel development of implantable devices that test a number of chemotherapy drugs simultaneously within the tumor in the living patient and permit readout of results so that the “winner” can be chosen. Even combinations can be preprogrammed and assessed.
Individualized drug sensitivity testing. Two different devices are illustrated that allow for in vivo drug sensitivity testing and biomarker analysis in patient tumors. (A) The Jonas et al. device is implantable and can test up to 16 different drugs simultaneously. (B) The Klinghoffer et al. device called CIVO microinjects up to eight different drugs or drug combinations before tumor sampling and has been tested in rodent and canine models and in human subjects. Various pharmacologic and pharmacodynamic markers were evaluated in both studies to demonstrate that the results correlated with clinical response to therapy. Reprinted with permission from Coombes RC, et al. Drug testing in the patient: toward personalized cancer treatment. Sci Transl Med. 2015:7;284ps10.
Individualized drug sensitivity testing. Two different devices are illustrated that allow for in vivo drug sensitivity testing and biomarker analysis in patient tumors. (A) The Jonas et al. device is implantable and can test up to 16 different drugs simultaneously. (B) The Klinghoffer et al. device called CIVO microinjects up to eight different drugs or drug combinations before tumor sampling and has been tested in rodent and canine models and in human subjects. Various pharmacologic and pharmacodynamic markers were evaluated in both studies to demonstrate that the results correlated with clinical response to therapy. Reprinted with permission from Coombes RC, et al. Drug testing in the patient: toward personalized cancer treatment. Sci Transl Med. 2015:7;284ps10.
Dr. Richard Klinghoffer and colleagues at Presage Biosciences, at the Fred Hutchinson Cancer Research Center, at Seattle Children’s Hospital, and at other sites developed a CIVO — a patented device that is inserted percutaneously into the tumor in a study subject in vivo. It permits evaluation of up to eight drugs or drug combinations at a time. The CIVO has an array of four to eight microinjectors for delivery (Figure). Tumor sections that are 4 µm thick at 2-mm intervals perpendicular to the injection column are examined to assess tumor response. Viable cells are marked blue with DAPI to label nuclear DNA, green with a tracking dye, and orange to identify a biomarker. Antimitotic activity of drugs is visualized by staining for phosphohistone H3, apoptosis by staining for cleaved caspase 3, and DNA damage by staining for phosphohistone H2AX. They also examined a drug-resistant tumor in their preclinical model to find an active drug among the 97 approved-drug panel from the National Cancer Institute, and they were able to identify that mTOR inhibitors had activity by observing decreased phosphorylation of the mTORC1 substrate of eIF4E-binding protein 1 (the “biomarker” for this drug). They were able to document images of every cell from each stained tissue using digital, automated, high-resolution whole-tissue scanning. The investigators reported that the results obtained by testing in a mouse tumor xenograft model correlated with results obtained by systemic drug administration. Data for four lymphoma patients tested in a pilot study were also presented, and the major side effect noted was mild grade 1 transient erythema.
Dr. Oliver Jonas and colleagues at Massachusetts Institute of Technology and their collaborators developed a different in vivo drug testing device. They deliver the 820-µm (diameter) × 3-mm (length) device via biopsy needle, and it remains in place for 24 hours. The device is a reservoir (Figure) that releases microdoses of drugs inside the tumor. A coring needle removes the relevant tissue from the tumor. Imaging of autofluorescent chemotherapy drugs such as doxorubicin, lapatinib, and sunitinib, or fluorescently labeled drugs, was by fluorescence microscopy; nonfluorescent drugs were visualized by matrix-assisted laser desorption/ionization mass spectrometry. The extracted sample is analyzed histologically for cleaved caspase 3 to identify apoptosis, Ki67 to mark proliferation, and survivin, an apoptosis inhibitor. The investigators validated that the same profiles of apoptosis were obtained whether the drugs were delivered intravenously or from the device.
In Brief
The technological limitations of these promising devices include site-to-site variation within a tumor or between one site and a distant metastasis, and genomic instability that leads to acquisition of drug resistance over time by selective pressure after treatment. Moreover, the major challenges for future development of this technology will be its cost and widespread access to millions of patients with cancer. However, the current system of drug development via a series of sequential phase I, phase II, phase III trials is fraught with exorbitant cost, time, and potential patient morbidity. The future will undoubtedly be revealed in next-generation prototypes of these devices.
