The rapid expansion of biomarkers, especially in the field of oncology, has created great excitement and hope among patients, as well as clinicians. As discussed in the main article in this publication, the use of biomarkers has revolutionized the treatment of many types of cancer in a very personalized way, and has significantly affected the entire cancer care team, including (but not limited to) oncologists, oncology nurses, oncology nurse navigators, surgical and clinical pathologists, and pharmacists. Therefore, it is imperative that healthcare professionals understand how biomarker results are obtained, their strengths and limitations, and the expectations that surround a given result.
It is essential that members of the cancer care team understand the significant difference between prognostic biomarkers and predictive biomarkers. A prognostic biomarker provides independent information regarding the likelihood of a clinical outcome (in the absence of treatment), such as disease progression. Conversely, a predictive biomarker provides information about the likelihood of response or nonresponse to a particular personalized treatment.1
Some biomarkers, such as HER2, may be both prognostic and predictive. For example, untreated HER2-positive breast cancers have worse outcomes than untreated HER2-negative breast cancers (prognostic); HER2-positive breast cancers respond to HER2-directed therapy, such as trastuzumab (predictive). A better understanding of which biomarkers predict a better or worse disease course or response to therapy will directly influence the choice of therapy, especially in a patient with multiple comorbidities. Similarly, an awareness of which biomarkers are appropriate for a specific malignancy in a specific patient can help to guide therapeutic decision-making and anticipate treatment-related toxicities (ie, personalized medicine).
However, members of the cancer care team need to understand much more than whether a biomarker is prognostic or predictive; they must also be aware of how a biomarker result is obtained. This knowledge is crucial to the interpretation of the result and confidence in that interpretation. In many cases, patients with documented advanced-stage cancer are no longer subjected to morbid surgical resections that typically provided a generous sample of their tumor. Instead, they are able to undergo minimally invasive procedures, such as computed tomographic- or ultrasound-guided fine-needle aspiration or core needle biopsies. These procedures usually yield only a small sample, which must be used to perform an increasing number of possible analyses. It is essential that the designated member of the cancer care team relay to the physician obtaining the sample (eg, the interventional radiologist or the minimally invasive surgeon) the need for adequate tissue for diagnosis as well as ancillary studies (eg, immunohistochemistry, flow immunophenotyping, fluorescence in situ hybridization, reverse transcription polymerase chain reaction, and next-generation sequencing).
The pathologist handling the tissue must be able to anticipate the need for possible future studies. For example, if the mass is an adenocarcinoma, adequate tissue for immunohistochemistry studies to rule in a lung adenocarcinoma and rule out a metastasis may be necessary; certain biomarkers, such as EGFR, ALK, and ROS1, may also be required by the oncologist. This “predicting the future” workup can be challenging in the event of a solitary cavitary lung mass, which may need all of the aforementioned studies or just culture for Mycobacterium tuberculosis. This is not an uncommon scenario in my practice, as the population we serve has a relatively high smoking rate and a relatively high M tuberculosis rate compared with other areas in the United States.
It is also critically important that the interventional radiologist, minimally invasive surgeon, and pathologist know how to handle the specimen properly, because improper handling, inadequate fixation, and other preanalytic variables may have a significant impact on the accuracy of the biomarker result.2 The cancer care team must always be aware of the sampling issues raised by these small biopsies, and have a low threshold for repeat biopsy if a laboratory result does not correlate clinically and/or radiographically.
Furthermore, in today’s cost-conscious environment it is imperative that members of the cancer care team understand which biomarkers are relevant to the malignancy that the patient harbors. Activating mutations in KRAS are recognized as strong predictors of resistance to EGFR-targeted therapy in metastatic colorectal cancer; whereas EGFR mutations in lung cancer may predict response to EGFR-targeted therapy (particularly tyrosine kinase inhibitors).3 Different programmed death ligand-1 (PD-L1) therapies are associated with different antibody clones used to perform the immunohistochemical studies that predict response.4 At our institution, we ask the oncology team which therapeutic is being considered before performing a PD-L1 biomarker analysis, to save financial resources as well as to ensure relevant testing and prevent misinterpretation.
When it comes to biomarkers, the members of the cancer care team play an essential role in obtaining the right tissue samples for the right tests at the right time, and overseeing the correct handling of those samples. These are critical ingredients for making medical oncology personalized.
- Nalejska E, Mączyńska E, Lewandowska MA. Prognostic and predictive biomarkers: tools in personalized oncology. Mol Diagn Ther. 2014;18:273-284.
- Yamashita-Kashima Y, Shu S, Yorozu K, et al. Importance of formalin fixing conditions for HER2 testing in gastric cancer: immunohistochemical staining and fluorescence in situ hybridization. Gastric Cancer. 2014;17:638-647.
- da Cunha Santos G, Shepherd FA, Tsao MS. EGFR mutation and lung cancer. Annu Rev Pathol. 2011;6:49-69.
- Hendry S, Byrne DJ, Wright GM, et al. Comparison of four PD-L1 immunohistochemical assays in lung cancer. J Thorac Oncol. 2018;13:367-376.