March 1, 2014 (Vol. 34, No. 5)
Frederick W. Wiese III, Ph.D. Manager EMD Millipore
Debra MacIvor, Ph.D. Product Manager EMD Millipore
From Novel Biomarkers to Combinatorial Therapies
As 2013 ended, it seemed that CAR (chimeric antigen receptor) T cell-based therapies had completely monopolized the limelight in the oncology theater. Electrifying data showed that most patients with leukemias and lymphomas treated with CAR T cells showed sustained or complete responses, and many remained cancer free. Carl June’s team even showed some promising results with CAR T cells in solid tumors. It’s not surprising that Novartis is now building an immunotherapy processing plant for scaling up chimeric T-cell production.
Can CAR T cells succeed universally where other therapies fail?
Only if they can overcome cancer’s hideous strength: heterogeneity. Cancer cells are endlessly changing. They are evading immune surveillance by secreting certain cytokines, they are escaping apoptosis, and they are thwarting antiproliferative cues. And within the increased local heterogeneity of solid tumors, cancer cells are ensconced in a microenvironment that inhibits T-cell migration. These solid tumors are tough targets—even for most CAR T cells. Attempts to specifically target tumor cells within this heterogeneous environment can result in toxicity.
It’s time to take a step back, then, and tackle heterogeneity in 2014, using a multipronged approach.
The problem of heterogeneity has long been appreciated among oncologists. Not only has intratumor genetic heterogeneity confounded targeted therapies, but intrapatient heterogeneity has also challenged interpretation of clinical trials. Advances in biomarker detection (protein, nucleic acid, metabolites, imaging signals), in combination with projects like The Cancer Genome Atlas (TCGA), Broad-Novartis Cancer Cell Line Encyclopedia, and patient-derived xenograft collections are helping cancer researchers quantify heterogeneity and determine its biological roles to shape therapeutic strategies.
The Impact of Genomic Heterogeneity on Cancer Drug Development
With the true “$1,000 genome” supposedly at hand, 2014 will witness increased whole-genome sequencing for tumors, their microenvironments, and normal tissues, along with novel approaches for parsing the data. TCGA, among other sequencing efforts, showed that baseline genomic heterogeneity makes it is difficult to establish a “normal” genetic background against which oncogenes and tumor suppressors can be identified.1 Modeling shows that sequencing up to 5,000 tumors per tumor type2 may be required to reveal all “driver mutations.”
This year will shed further light on sources of heterogeneity, such as epigenomic alterations,3 splicing variations,4 and changes in DNA repair.5 Technical breakthroughs, such as sensitive assays for epigenetic marks, high-throughput functional assays, and advances in RNA analysis—in combination with integration of patients’ genetic and clinical data, such as the MD Anderson Cancer Center’s “APOLLO” program—may reveal the contributions of these factors to tumor progression and response to therapy.
Using Insights into Intratumor Heterogeneity to Direct Treatment
In 2014, we are likely to make progress in quantifying the link between intratumor heterogeneity and drug response using xenograft libraries, engineered mouse models, and computational tools. The expansion in sequencing capabilities has revealed that the mutations in a tumor that enable drug resistance, relapse or metastasis are present (at low frequencies) in the primary tumor.
Multiple sampling from patients enables detection of changes in a tumor’s genome over time (temporal heterogeneity) and that the degree of these changes may be an accurate prognostic indicator.6 Computational modeling of heterogeneity, assuming that tumors are composed of treatment-sensitive and treatment-resistant cells, may start being used to direct dosing schedules for chemotherapy or radiation, potentially determining therapeutic success or failure.7
Combating Heterogeneity with Biomarker-Directed, Targeted Therapies
Given the heterogeneity of a patient population, measuring the right biomarkers can help demonstrate efficacy of a targeted therapy. Thanks to new biomarkers, careful trial design, and target choices, we are likely to see increased success of targeted therapies in 2014 compared to recent years.
Progress will continue in drugging novel target classes: epigenetic regulators, immune checkpoint proteins, and metabolic enzymes. Using noninvasive quantification of the biomarker 2-hydroxyglutarate in AML patients, the effectiveness of mutant isocitrate dehydrogenase inhibitor AGI-221 (clinical trial ongoing) may be more effectively monitored. And, though limited by heterogeneous expression and lack of an accurate assay, PD-L1 expression has shown to be a promising biomarker for predicting response to the immunomodulator nivolumab, now in Phase III clinical trials. Finally, epigenetic modulators started the year strong, with the protein methyltransferase inhibitor EPZ-5676 showing promise in patients with hematologic malignancies bearing MLL translocations.
Biomarker-driven studies may also facilitate progress for the hypertargeted drug candidates revealed in 2013, such as the proposed specific inhibitor of oncogenic Kras8 and AZD-9291, an irreversible inhibitor of mutant EGFR receptors that have acquired the T790M resistance mutation.
Overcoming Challenges Facing Biomarker Measurement
Can we really attack heterogeneity by simply measuring more biomarkers, in more patients, at more time points? Among the obstacles to sampling is patient consent, given that multiple samples must be taken and used for yet-unforeseen purposes. Sample amounts are still limiting, though technical advances now allow simultaneous measurement of multiple analytes of the same type (i.e., circulating proteins), and even simultaneous measurement of RNA and protein (Figure).
In 2014, we will see more information extracted from liquid biopsies, such as imaging cytometry of circulating tumor cells or measurement of circulating protein and nucleic acid biomarkers. Because liquid biopsy volumes are limited, we’re likely to see a wider adoption of microscale and nanoscale assays and new tracers for noninvasive, functional imaging.
These advances in assay technology should also spur discussions on how to collect, prepare, and preserve samples while maintaining their nucleic acid, protein, and morphological profiles. Developments such as imaging mass spectrometry, on-line microscale sample preparation coupled to liquid chromatography, and simultaneous, hands-free protein purification and concentration protocols may see more use in bioanalysis, maximizing information obtained while minimizing sample handling/perturbation.9,10
Combinatorial Therapies: Bringing in Biologics
A 2014 forecast would not be complete without considering new combinatorial therapies for cancer. The year began with an accelerated approval for treating certain metastatic melanomas with a combination of two molecules (trametinib and dabrafenib) that target the MEK and BRAF kinases.
However, unlike most kinase inhibitors, many of today’s drugs are large molecule- or cell-based biotherapeutics such as monoclonal antibodies (mAbs), cancer vaccines, and adoptive immunotherapies, and combining these drugs poses unique regulatory and therapeutic challenges. In the following months, we may see reports from more trials currently under way for biotherapeutic combinations: nivolumab plus ipilimumab (dual blockade of two immune system checkpoints), nivolumab plus an anti-LAG3 antibody, bevacizumab plus ipilimumab (angiogenesis inhibition with immunomodulation), and vemurafenib plus ipilimumab (BRAF kinase inhibition plus immunomodulation).
Regulatory agencies do recognize that combining biotherapeutics holds enormous potential for treating even the most heterogeneous and refractory of cancers. In 2013, the FDA approved the combination of pertuzumab and trastuzumab, two HER2-specific mAbs, for treating breast cancer in combination with chemotherapy. However, challenges to combining biotherapeutics include differential dosing and effects on immunogenicity, toxicity, and pharmacokinetics/pharmacodyamics, not to mention the challenges of regulating co-manufacture and co-packaging. Furthermore, traditional endpoints used to measure efficacy may not be sufficient to determine whether combinations of biotherapeutics, especially immunotherapies, are efficacious.
Ke Liu of the U.S. FDA, in a panel discussion at October’s Molecular Targets and Cancer Therapeutics conference, proposed applying the guidance regarding combinations of small molecule drugs to combinations of biologics and working with drug developers to establish better biomarkers for toxicity, PK/PD, and immunotherapeutic efficacy. We hope that 2014 brings progress toward some of these goals, bringing new combinatorial biotherapeutics closer to widespread clinical use.
1 Nature. 2013 Jul 11;499(7457):214-8.
2 Nature. 2014 Jan 5. doi:10.1038/nature12912. [Epub ahead of print]
3 Science. 2013 Mar 29;339(6127):1567-70.
4 Clin Cancer Res. 2013 Nov 27. [Epub ahead of print]
5 PLoS One. 2013 Dec 12;8(12):e83527.doi: 10.1371/journal.pone.0083527.
6 F. Markowetz and colleagues, unpublished.
7 Leder K* et al. Cell, in press.
8 Nature. 2013 Nov 28;503(7477):548-51.
9 Proteomics Clin Appl. 2013 Dec;7(11-12):733-8.
10 J Pharm Biomed Anal. 2014 Jan;87:120-9.