Emerging laboratory technology will be “moving the bar forward” in terms of molecular markers, genomics, and gene-expression profiling, with the potential for huge payoffs to oncologists and patients, according to Mark Pegram, MD, the Susy Yuan-Huey Hung Professor of Medicine at Stanford School of Medicine, Director of the Breast Cancer Oncology Program at Stanford Women’s Cancer Center, and Co-Director of Stanford’s Translational Oncology Program in Palo Alto, California.
At the 10th Annual New Orleans Summer Cancer Meeting, Dr. Pegram described novel approaches as they might be applied in breast cancer, showing enthusiasm for their possibilities, tempered with a note of caution.1
“These are technologies emerging in the lab, and what the future may hold for their clinical application, we don’t yet know,” he said. “Hopefully, in the end, these will be useful for clinical decision-making and result in improved patient outcomes.”
Dr. Pegram also described the challenge of developing targeted therapies for rare mutations, even with these sophisticated tools.
End of the Pathologist?
“At Stanford, surgical oncology has conspired to try to eliminate the need for the pathologist,” Dr. Pegram quipped.
Virtually all the information useful to oncologists comes by way of the pathologist’s review—an established practice that might be disrupted by cutting-edge technology called selective negative ion mode desorption electrospray ionization mass spectrometry, developed at Stanford, he said. Desorption electrospray ionization mass spectrometry is an ionization technique that can image biologic samples without the need for extensive sample preparation, investigating the distribution of diagnostic lipids and metabolites directly from tiny tissue sections.
As the developers explained in a published investigation,2 “Samples are bombarded with microdroplets that dissolve hundreds of lipids and metabolites. The splash forms secondary microdroplets that enter a mass spectrometer, providing a detailed chemical map of the distribution of molecules within the sample surface.”
Dr. Pegram added, “On a thin sample of tumor tissue you can measure, based on their molecular weight, all the biologic elements that splash out of that tumor sample. You find lipids, phospholipids, proteins, products of metabolism—everything that’s in there.”
Molecular masses of these multiple types of molecules from the tumor are compared to normal tissue and used to construct patterns that differentiate the tumor from the epithelium and stroma. The technology may someday provide a means of interrogating tumor tissue without the need for a large surgical sample or reading by pathologist, he said.
Where Next-Generation Sequencing Falls Short
Outside of a handful of breast cancer–associated genes with frequencies greater than 10%, “the sobering reality” is that most other mutations are “one-offs” that are unique to individuals or small groups of individuals, Dr. Pegram noted. Next-generation sequencing may be able to identify these new targets, but developing drugs to target them will be difficult.
The HER2 kinase mutation is an example. This mutation is found in patients with HER2-negative breast cancer, especially of the lobular subtype. It occurs in the absence of gene amplification and has a frequency of only 1.6% (3% in lobular tumors).
“Because these mutations tend to occur in the kinase domain, it sounds reasonable to treat with lapatinib [Tykerb]; however, this doesn’t work,” he indicated. “These recombinant constructs of mutants do not bind to lapatinib anymore, but they do bind to the tyrosine kinase inhibitor neratinib.”
In a collaboration spearheaded by the group at Washington University in St. Louis, Dr. Pegram and colleagues are studying neratinib in a small cohort of patients with HER2 mutations. Taking into account the fraction of patients who harbor a HER2 kinase mutation, meet all eligibility criteria, and are willing to participate in a clinical trial, the researchers calculated the “number needed to study” neratinib in this population, concluding that in the best-case scenario, 125 patients must be screened to yield 1 enrollee. Nevertheless, the study did launch, with a low double-digit number of patients currently enrolled.
“One response to neratinib, a drug we thought would have exquisite activity,” Dr. Pegram said. This is “sobering” if it foreshadows the payoff for such endeavors, he commented.
Tackling Heterogeneity
“Another conundrum in tumor genome sequencing campaigns is the degree of heterogeneity in breast cancer,” he continued. Genomes sequenced for more than 100-fold coverage will identify separate clones within the same primary tumor that have different genotypes—and this can be observed early, at diagnosis.
One of the first examples of this complex heterogeneity was a discovery made in renal cell carcinoma. Gerlinger and colleagues evaluated spatially separated samples from the primary tumor and associated metastases using exome sequencing.3 A single biopsy showed about 70 mutations that accounted for only half of all the mutations identified across multiple biopsies. Only 34% were contained in all regions.
“Breast cancer is the same,” Dr. Pegram indicated. “We are wrestling with how to accommodate this heterogeneity.”
Harnessing the Dynamic Nature of Mutations
Moreover, these sequences are ever-changing as a consequence of selection pressure after treatment. It may be possible, however, to turn this characteristic into a positive feature.
Vanderbilt investigators studied the molecular landscape of tumors after neoadjuvant chemotherapy in patients with triple-negative breast cancer, focusing on the molecular underpinnings of patients lacking a complete pathologic response (a possible marker of resistance). By next-generation sequencing and digital RNA expression analysis, they identified diverse molecular lesions and pathway activation in drug-resistant tumor cells, compared to pretreatment biopsies; they hypothesized that these findings may mirror micrometastasis destined to recur clinically. Ninety percent of the tumors contained a genetic alteration potentially treatable with available targeted drugs.4
“Tumors were enriched for mutations along pathways we all know—those related to DNA repair, growth factor receptor amplification, PI3K/AKT, Ras/RAF/ERK, and cell cycle,”
Dr. Pegram noted. “This suggests that with this type of analysis we may have the opportunity for personalized adjuvant treatment.”
Mutations of the estrogen receptor are further examples of the dynamic nature of mutations. Estrogen receptor mutations occur in up to 20% of metastatic tumors (vs about 1% in primary tumors). The frequency appears to be enriched with prior endocrine treatment, and there is a clear trend toward rising mutation frequency from primary disease to early metastatic disease to late metastatic disease.
“These mutations are also quite interesting,” he observed. “They cluster in the ligand-binding domain, which begs the question of whether antiestrogen therapeutics will hit them…. This could prove to be an opportunity, now that we understand that estrogen receptor mutations are not uncommon.”
‘Composite’ Look at Heterogeneity
Could circulating tumor cells provide a snapshot of the tumor’s global picture? A new cell separation method that employs a microfluidic device yields live tumor cells that can not only be enumerated but analyzed for their components, such as steroid receptor, HER2, mutations, and proteins.
“In theory, this could be powerful,” Dr. Pegram commented. “We could measure the percentage of cells that are [estrogen receptor]–positive in the circulation, and if it reflects the global tumor burden, we can estimate the degree of heterogeneity of that marker and perhaps correlate that with the expectation of response to an antiestrogen, for example.”
It is actually now possible to isolate single cells and observe their makeup as well. In a collaboration with Amy Herr, PhD, of the Department of Bioengineering, University of California at Berkeley, who has pioneered a new technology, Dr. Pegram’s lab is doing this, focusing on the protein content of circulating tumor cells. Using a single-cell Western blot analysis platform, researchers place individual cells into microwells, where they can be lysed and electrophoresed, producing a readout of all proteins within that cell.5
Single cells can also be subjected to whole-exome sequencing. This may better capture “what’s in the general mix” of the tumor, as compared to a biopsy of just one site. This approach captures cell-to-cell heterogeneity; provides objective, quantitative, reproducible, digital data for archiving; and “spares precious patient tissue,” Dr. Pegram said. “Maybe this is a path forward toward getting a handle on the heterogeneity of circulating tumor cells.”
Moreover, cells may not be needed at all to produce a global snapshot, as tumor-specific DNA can now be isolated from the circulation. This cell-free DNA approach can correlate the amount of circulating DNA with radiographic findings, can quantitate second-site mutations, and can be applied to any tumor type. “This could be very interesting in terms of following patients with metastatic disease,” he said.
Dr. Pegram concluded by acknowledging that the path to the development of new biomarkers is long and arduous but could ultimately lead to better treatments and outcomes. ■
Disclosure: Dr. Pegram reported no potential conflicts of interest.
References
1. Pegram M: Novel molecular markers, genomics, and gene expression profile: Moving forward the bar for breast cancer therapy development. 2015 New Orleans Summer Cancer Meeting. Presented July 18, 2015.
2. Eberlin LS, Gabay M, Fan AC, et al: Alteration of the lipid profile in lymphomas induced by MYC overexpression. Proc Natl Acad Sci USA 111:10450-10455, 2014.
3. Gerlinger M, Rowan AJ, Horswell S, et al: Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 366:883-892, 2012.
4. Balko JM, Giltnane JM, Wang K, et al: Molecular profiling of the residual disease of triple-negative breast cancer after neoadjuvant chemotherapy identifies actionable therapeutic targets. Cancer Discov 4:232-245, 2014.
5. Hughes AJ, Spelke DP, Xu Z, et al: Single-cell western blotting. Nat Methods 11:749-755, 2014.
