The current generation of oncologists has witnessed great advances in our understanding of tumor biology and biomarkers linked to treatments. Those advances started with research, but disseminating this information can be difficult given the myriad of obstacles in adoption to practice. The science behind these advances is fascinating and excites those in medicine with the possibility of providing meaningful, life-altering care to patients. But still there exists the reality of the vetting of each new discovery, starting with niche use among the early users, before it gets adopted more broadly.
With the advent of molecular medicine, we have novel options to detect and identify genetic mutations and other biomarkers to assist in selecting the appropriate therapy to target cancer cells. Combining this knowledge with guidelines of how to treat based on tumor type can only serve to improve patient outcomes.
There are many options in selecting tests to gather pertinent information related to a patient’s genetic profile and the biology of their tumor. In this current installment of Interview with the Innovators, and with the intent to assist in the dissemination of impactful information, we focus on the bioT3 approach, which provides a molecular diagnosis for tumors with unclear diagnosis as well as comprehensive biomarker profiling, including mutational analysis and protein expression markers to assist oncologists in selecting site-specific and targeted therapy options for patients with metastatic cancer. The publishers of PMO had the pleasure of meeting with Dr Ralph V. Boccia from the Center for Cancer and Blood Disorders and Clinical Associate Professor at Georgetown University who participated in the research for these products and has first-hand experience with them in the clinic. To view the live interview, please visit www.PersonalizedMedOnc.com/videolibrary.
PMO Genomic tools such as next-generation sequencing are not widely adopted in community practices for metastatic patients. In your experience, what are the barriers to adoption?
Dr Boccia The low uptake in the community is primarily due to the fact that the first wave of tools was developed to meet the needs of academia and don’t adequately address the needs of community practices. More specifically, when it comes to next-gen sequencing, research-focused oncologists are interested in understanding all mutations associated with a given tumor type, even if no agents are currently available that can target the related pathways. Increasingly, there is interest among these academics in obtaining information across the entire genome in the hope of future actionability, and often in the context of their collaborations with research divisions of pharmaceutical companies.
In sharp contrast, treatment-focused oncologists are primarily interested in current actionability and thus interested in biomarkers linked to Food and Drug Administration (FDA)-approved drugs and late-stage clinical trial candidates. Further, they are focused on getting the information fast in a simple and easy-to-understand tumor-specific report at a reasonable cost. Therefore, the use will increase when the needs of community practices are more directly addressed.
PMO Can you discuss the strengths and weaknesses of the various next-gen sequencing tests to detect actionable biomarkers?
Dr Boccia In terms of the first wave of tools developed for academia, there are 2 main approaches: 1) sequencing only, and 2) testing everything independent of tumor type. The strength of the “sequencing-only” approach is that it consumes very little tissue but, unfortunately, it is not a comprehensive platform. For some specific biomarkers, IHC [immunohistochemistry] and FISH [fluorescence in situ hybridization] are the more appropriate platforms versus sequencing. IHC probes protein expression levels in the tumor, which cannot be performed with sequencing. This is relevant both for traditionally important biomarkers such as HER2, ER [estrogen receptor], and PR [progesterone receptor] expression, which are recognized to be important in breast cancer, as well as relatively new biomarkers such as programmed death-ligand 1 (PD-L1), which may have a growing clinical relevance in melanoma, lung cancer, and other solid tumors.
Sequencing also probes genetic rearrangements and amplifications in a manner that is inconsistent with approved targeted therapy labels and inclusion criteria of many clinical trials. Recent ASCO/CAP [American Society of Clinical Oncology/College of American Pathologists] guidelines for biomarker testing in lung cancer clearly state that ALK rearrangement testing should be done by FISH, in line with the pivotal studies for crizotinib and the corresponding package insert. Rearrangements, in principle, can also be probed by sequencing, but the cutoff criteria and the test specifics are not identical. Unlike academia, where the remaining tests can be done by in-house pathology, it is not pragmatic for a community oncologist to work with multiple testing facilities to obtain comprehensive biomarker information.
The “testing everything independent of tumor type” approach employs a very large number of biomarker tests. The strength of this approach is that it is extremely comprehensive. However, when you run such a large number of biomarkers independent of the type of cancer, you end up with slower turnaround times, high costs, and significant tissue use, in conflict with the basic needs for a community oncology practice. In addition, many biomarkers are only relevant within specific tumor types, so this approach is not aligned with the clinical evidence behind many biomarkers. High-cost tests like this are better used as a last resort when a patient has exhausted all standard-of-care options.
PMO How common is the issue of unclear diagnosis in metastatic cancer?
Dr Boccia In clinical practice, patients can present with carcinomas for which we cannot identify their primary site—otherwise called carcinomas of unknown primary site. In addition, there are certain ambiguous presentations where patients appear to have a certain type of tumor; however, the overlap of the clinical presentation with potentially different types of cancer makes the initial diagnosis unclear.
Data would suggest that the combination of carcinomas of unknown primary site and the ambiguous presentations probably make up for around 100,000 patients newly diagnosed each year, which is ~15% of all newly diagnosed metastatic patients. So it’s no small number.
It is important for clinicians to identify the primary site, because the treatment that we render is dictated by where the tumor started, not where the tumor ends up. Accurate diagnosis of tumor type/subtype is necessary for selecting the proper site-specific chemotherapy and molecularly targeted therapies. It is important to note that molecularly targeted therapies are indicated for specific tumor types. For example, vemurafenib is FDA approved for metastatic melanoma patients harboring a BRAF V600E mutation, but it is not approved in metastatic colorectal cancer with the same mutation and may not be effective in this case. So understanding the biomarker profile in the context of tumor type is important.
PMO We understand that you have used bioT3 from bioTheranostics in your clinical practice. Can you share with us the type of patient for whom you use this approach, and how it compares to other offerings that you are familiar with?
Dr Boccia I’ve had the opportunity to work with bioTheranostics for a number of years, first in the research setting and now in clinical practice. bioT3 was not commercially available when I first started doing research with them as part of a consortium looking at carcinomas of unknown primary site. I currently use bioT3 in my everyday practice for metastatic patients with clear as well as unclear diagnoses, from initial treatment through resistance and recurrence.
The bioT3 offering combines tumor type diagnosis and comprehensive biomarker profiling for metastatic tumors. It is made up of 2 components, the first being CancerTYPE ID, a gene expression–based molecular tumor classifier. CancerTYPE ID is necessary when we’re looking to identify a primary tumor type, specifically if we’re confronted with a carcinoma of unknown primary site or potentially one of those ambiguous states where there is a differential diagnosis. An example of that might be metastatic squamous cell carcinoma in the lung that could be a primary bronchogenic squamous cell carcinoma in the lung, or it could be a primary head and neck squamous cell carcinoma metastatic to the lung. If unidentified using regular tools, CancerTYPE ID can help distinguish between these tumor types. Another example might be a patient who presents with abdominal carcinomatosis, and we don’t know if it’s a GI [gastrointestinal] or a GU [genitourinary] primary. CancerTYPE ID can be useful for cases in which the tumor type is unknown, and also in cases where there is diagnostic ambiguity and several tumor type possibilities exist.
The second component is CancerTREATMENT NGS+, a comprehensive platform that includes next-generation sequencing, FISH, and IHC that lends us the opportunity to select therapies. The biomarkers tested by CancerTREATMENT NGS+ are based on National Comprehensive Cancer Network and ASCO guidelines and major phase 2 and phase 3 clinical trials, resulting in more concise panels and reports. This is in contrast to the complicated reports you receive when you do extensive next-generation sequencing that leave a lot of physicians in a position where they don’t understand where to go next.
PMO How do you view the strength of the clinical validation evidence for the CancerTYPE ID, and how does it compare with other gene expression tests?
Dr Boccia In comparison to other gene expression tests, CancerTYPE ID covers a significantly larger number of tumor types, 50 tumor types compared with 15 and 42. In addition, several studies have validated its accuracy and clinical utility. Specifically, the Mayo Clinic, University of California Los Angeles, and Massachusetts General Hospital have done a blinded study and documented the accuracy of CancerTYPE ID (Clin Cancer Res. 2012;18:3952-3960). A head-to-head comparison of CancerTYPE ID versus IHC demonstrated that CancerTYPE ID was significantly better (J Mol Diagn. 2013;15:263-269). A prospective study in patients with carcinomas of unknown primary site that we published in the Journal of Clinical Oncology showed a 37% improvement in overall survival in those patients profiled with this assay and then treated based on the tumor type identified compared with empiric standard-of-care chemotherapy (J Clin Oncol. 2013;31:217-223).
PMO For what types of patients do you see a role for CancerTYPE ID in clinical practice?
Dr Boccia We’ve used CancerTYPE ID in our clinical practice in a number of situations. When we get a pathology report back, the first thing we do is to look to see how they document where the tumor that they’re describing appears to have originated from. Oftentimes there is a long account of immunohistochemical stains describing what they conclude. But sometimes pathology reports have an inconclusive or ambiguous diagnosis from a histologic and immunohistochemical standpoint. An example might be a patient with a suspected diagnosis of non–small cell lung cancer [NSCLC] based on clinical correlation, but the pathology report indicates that the tumor is TTF-1 [thyroid transcription factor-1] negative, whereas most lung cancers, but not all lung cancers, are TTF-1 positive. If we think it is NSCLC but can’t be absolutely certain, that would be a reason to use the CancerTYPE ID.
Additionally, if the clinical presentation is atypical, or if there are multiple lesions at a distant point from any organ, and standard workup does not provide a definitive diagnosis, then CancerTYPE ID can be very helpful in these situations.
PMO Considering the second component of bioT3, for which patients do you see a role for CancerTREATMENT NGS+ in your clinical practice?
Dr Boccia We find CancerTREATMENT NGS+ to be helpful in several situations. The first would be at the time of diagnosis where we’re looking to profile that patient’s tumor and make sure that we have in fact identified actionable targets.
An example of that might be NSCLC, the nonsquamous variety, where we want to be sure that we have given the patient the best treatment options since there are several targeted therapies available based on the biomarker profile of the tumor. If, for instance, the tumor is EGFR, ALK-1, or ROS1 positive, that’s not a patient we want to be giving chemotherapy to at the outset, because randomized trials have clearly shown that the tyrosine kinase inhibitors offer the patient better response rates, better progression-free survival, and better overall survival. On the other hand, if these biomarkers are negative, then this patient is best suited for chemotherapy. So patients with a known diagnosis of metastatic NSCLC are candidates for CancerTREATMENT NGS+ in the up-front setting. CancerTREATMENT NGS+ is a great platform because it combines next-gen sequencing with FISH and IHC testing, which is important for NSCLC if you are looking for ALK rearrangements because ASCO/CAP guidelines recommend that ALK testing be performed by FISH. If you’re looking for PD-L1 expression, the best way to assess this is through IHC.
CancerTREATMENT NGS+ is also very useful for those metastatic cases in which several lines of therapy have been implemented and the disease progresses. A good example of that would be a breast cancer patient with ER+/PR+ tumor type; we would offer several lines of endocrine therapy before moving on to chemotherapy. Once we’ve gotten beyond those first several lines of therapy and looking for additional actionable targets or clinical trial options, that would be a good time to use CancerTREATMENT NGS+ secondarily.
There are times when we would use the 2 components of bioT3 together. An example would be an unknown primary site, let’s say an ambiguous primary site or one that we thought might be lung. CancerTYPE ID would help us confirm this is an NSCLC, and that it is nonsquamous, to allow us to better understand whether there is an actionable target, specifically ALK, ROS1, RET, and EGFR. So combining them in this instance would be a perfect tool for us to set up a treatment program that we could carry through for many lines of therapy.
PMO An important theme at ASCO 2014 is the importance of managing costs of cancer care, particularly for metastatic patients. How important is it to lower cost from the perspective of community practices, and how does bioT3 address this critical need?
Dr Boccia We’re at a point in this society where healthcare costs are clearly out of control. The budget is unsustainable, and it’s important that all of us contribute to controlling costs the best we can. There is ongoing payment reform, and so the whole system is changing in the next several years. What this means is that we’re going to be sharing risks with the carriers, as well as showing value and quality with our treatment selection. Accountable care organizations are forming and will recruit members, and all will share risk. The upside potential here is enormous to begin to control the cost of some of these more expensive tests and provide actionable answers for more effective therapy. In looking at Explanation of Benefits coming in from my patients for the testing we have ordered, I’m sometimes appalled. I’ve seen bills for profiling as high as $30,000, certainly $5000, $6000, $7000, and $8000. As a treatment-focused community physician, it is very important for me to obtain actionable information in a cost-effective manner, and I think the bioT3 approach takes this into account.
PMO Thank you very much for your time today, and our best wishes to you for continued success.
Dr Boccia Thank you, it was my pleasure.
