Resolution of Diagnostic Dilemmas in Oncology


Diagnostic uncertainty occurs in 1 in 6 metastatic cancers1-3

Approximately 600,000 patients/year are diagnosed with metastatic cancer.1-3 Of these, ~100,000 display diagnostic uncertainty related to a spectrum of issues.1-3 Some differential diagnoses include lung squamous versus head and neck squamous cell carcinomas, mesothelioma or lung cancer, and subtyping specific neuroendocrine tumors.4,5 Identification of an accurate tumor type diagnosis remains important particularly when clinical presentation is atypical and/or immunohistochemistry (IHC) results do not clearly define a single diagnosis and treatment options can vary widely.

IHC fails to provide an accurate diagnosis in a significant number of metastatic cancers

Panels of IHC markers commonly used to determine the primary tumor type of metastatic cancers can yield ambiguous results.6,7 An MD Anderson Cancer Center study found diagnostic errors—major discrepancies that potentially affected patient care—in 6.2% of 2,718 cases retrospectively reviewed by expert pathology subspecialists.8 Minor discrepancies were found in an additional 18.7% of cases. In another study reported by Weiss et al and discussed in more detail below, about 1 in 3 metastatic cases were not accurately diagnosed.9

CancerTYPE ID®: A test addressing diagnostic uncertainty in metastatic cancer

CancerTYPE ID is a 92-gene expression assay that classifies tumors by matching the gene expression pattern of a patient’s tumor tissue to a database of known tumor types and histological subtypes covering more than 95% of all solid tumors based on incidence. The assay employs tumor enrichment by laser microdissection to spare tissue for testing and a sensitive a quantitative RT-PCR platform that requires only 300 non-necrotic tumor cells. The assay analyzes the expression of lineage-specific transcription factors, plasma membrane proteins, and other tissue-specific tumor markers to distinguish between 50 tumor types and subtypes. Since its launch in 2008, > 6,000 oncologists and > 5,000 pathologists have used CancerTYPE ID.10 To date, more than 23,000 patient tumor samples have been analyzed.11

CancerTYPE ID can be effective in providing a tumor type diagnosis where conventional histopathology methods have provided an uncertain diagnosis and two or more tumor types are suspected, as well as in other challenging diagnostic situations. Because of its ability to provide additional information on histological subtypes, CancerTYPE ID has shown particular clinical utility in differentiating between subtypes in difficult cases of lung and gastrointestinal carcinomas, as well as with subtyping of different neuroendocrine tumors.12 Furthermore, the CancerTYPE ID workflow, which involves tumor enrichment and a multiplexed RT-PCR assay has been optimized for high performance in cases where there is limited tissue available for analysis.13

CancerTYPE ID clinical validation

CancerTYPE ID was validated in a prospective, blinded, multi-site clinical trial involving 790 samples consisting of 50 tumor types/subtypes that included both metastatic and primary tumors types with a range of tumor grades, and also included limited tissue samples. In this study, CancerTYPE ID showed 87% overall accuracy for tumor type. 14

A subset analysis of tumors for which limited tissue samples were available (n=109) CancerTYPE ID demonstrated 91% overall accuracy.13 This finding with limited tissue samples is especially valuable with trends toward the use of minimally invasive techniques for tissue biopsy.

A second pivotal study compared the accuracy of CancerTYPE ID vs IHC/morphology for diagnosis of clinically challenging primarily metastatic tumors.9 Archived samples (N=122) were analyzed via both CancerTYPE ID and IHC/morphology and results compared to reference diagnosis of tumor type based on clinical information and pathologic workup. In this study, CancerTYPE ID demonstrated overall accuracy of 79%—significantly higher (P=0.019) than the 69% achieved by IHC/morphology (Figure 1). This finding suggested to the authors that poorly differentiated tumors may retain their RNA profile to a greater extent than their morphologic and protein profile. Note the performance of IHC/morphology relative to CancerTYPE ID declined significantly when greater than 7-9 stains were used for evaluation.

Figure 1: Diagnostic accuracy: CancerTYPE ID vs IHC/morphology

A prospective outcomes study was conducted in 289 previously untreated patients with a cancer of unknown primary (CUP) diagnosis.15 Successful assays were performed in 252 patients, of these 247 (98%), had a tissue of origin predicted by CancerTYPE ID. CancerTYPE ID-directed, site-specific treatment was provided to 194 patients. Overall median survival time was 12.5 months for these 194 patients vs 9.1 months for a historical control group of CUP patients treated empirically in previous CUP trials; a 37% difference. In addition, when CancerTYPE ID predicted tumor types that were clinically more responsive, median survival was significantly improved vs predictions of less responsive tumors (13.4 vs 7.6 months; P=0.04).

Clinical utility has also been demonstrated in the real-world practice environment. A 28-site prospective, multidisciplinary study involved 73 medical oncologists and 34 pathologists, with 397 cases included in the final analysis. Use of the CancerTYPE ID 92-gene assay changed the initial treatment decision in 46% of cases.17 In a retrospective survey-based study, 103 participating physicians, mostly in community-based practices (82%) completed a web-based survey. Patients underwent extensive diagnostic testing prior to the physicians ordering the CANCERTYPE ID assay. The addition of CancerTYPE ID testing enabled the identification of new cancers not previously considered in about one quarter of cases (28% from survey data).18

CancerTYPE ID: A validated tool that can help limit diagnostic ambiguity and aid in clinical decision-making

In sum, many metastatic cancers involve some degree of diagnostic ambiguity. It’s important to clear up that ambiguity where possible, especially in this day and age of targeted therapies and immunotherapies. By providing diagnostic certainty in many cases, CancerTYPE ID can potentially aid in the delivery of precision medicine for metastatic cancer including site-specific chemotherapy, molecular targeted therapy, or immunotherapy.


Want more information? Order a test? Start here.


Thank you for your interest in CancerTYPE ID.



  2. Thomas SP et al. J Clin Oncol. 2015;33(suppl 3; abstr 249).
  3. Pavlidis N et al. Eur J Cancer 2003;39:1990
  4. Schroeder BE et al. Mod Pathol. 2012;25.
  5. Thomas SP et al. J Clin Oncol. 2015;33(suppl 3; abstr 15206).
  6. Conner JR, Hornick JL. Adv Anat Pathol. 2015;22(3):149-67.
  7. Anderson GG, Weiss LM. Appl Immunohistochem Mol Morphol. 2010;18(1):3-8.
  8. Middleton LP et al. J Oncol Pract. 2014;10(4):275-80.
  9. Weiss LM et al. J Mol Diagn. 2013;15(2):263-9.
  10. Source: CancerTYPE ID website/Healthcare Professionals:
  11. Data on file at Biotheranostics, Inc.
  12. Kerr SE et al. Mod Pathol. 2014;27(1):44-54.
    Advance Online: Mod Pathol. (2013) 1-11.
  13. Brachtel EF et al. Oncotarget. 2016;7(19):27220-31. doi: 10.18632/oncotarget.8449.
  14. Kerr SE et al. Clin Cancer Res. 2012;18(14):3952-60.
  15. Hainsworth JD et al. J Clin Oncol. 2013;31(2):217-23.
  16. Greco FA et al. J Clin Oncol. 2016;34(suppl; abstr 11554).
  17. Thomas SE, et al. J Clin Oncol 2015;33(suppl 3; abstract 15206)
  18. Kim et al. Personalized Medicine Onc. 2013;2:68-76.
CancerTYPE ID Intended Use and Limitations
CancerTYPE ID is indicated for use in tumor specimens from patients diagnosed with malignant disease and is intended to aid in the classification of the tissue of origin and tumor subtype in conjunction with standard clinical and pathological assessment by a qualified physician. CancerTYPE ID is not intended to predict patient survival benefit, treatment efficacy or to distinguish between benign versus malignant lesions. Tumor types not included in the CancerTYPE ID reference database may exhibit RNA expression patterns that are similar to RNA expression patterns within the reference database. This test was developed and performance characteristics have been determined by Biotheranostics, Inc. It has not been cleared or approved by the U.S. Food and Drug Administration. This test is used for clinical purposes. It should not be regarded as investigational or for research. How this information is used to guide patient care is the responsibility of the physician. Biotheranostics is certified under the Clinical Laboratory Improvement Amendments of 1988 as qualified to perform high-complexity clinical laboratory testing.