Caris GPSai

Caris GPSai, a Genomic Probability Score, uses whole exome (DNA) sequencing and whole transcriptome (RNA) sequencing coupled with machine learning to aid in identifying the tissue of origin.

Identify Tumor Origin and Improve Treatment Decisions with Caris GPSai

Cancer of Unknown Primary (CUP) diagnosis is typically treated empirically and has very poor outcomes, with median overall survival less than one year.1 Caris GPSai™, a Genomic Probability Score, uses whole exome (DNA) and whole transcriptome (RNA) sequencing coupled with trained and validated neural networks to aid in identifying the tissue of origin.

Caris GPSai analyzes genomic and transcriptomic data to match a tumor’s molecular signature to one or more of 90 cancer categories from the Caris database. Caris GPSai provides additional insight to help oncologists better manage CUP or cases with atypical clinical presentation or clinical ambiguity.

Caris GPSai™ is included for all CUP cases and also may be selected on the requisition for other tumor types.

Enhance Diagnostic Confidence with Artificial Intelligence

Considering that rates of inaccurate diagnosis range between 3% and 9%2, Caris GPSai provides an integral part of quality control and may lead to improved diagnostic accuracy.

The Caris GPSai algorithm was trained and retrospectively validated on over 250,000 cases from the Caris database. The algorithm was then validated prospectively on an additional 3,200 MI Tumor Seek Hybrid™ cases.

VALIDATION STUDY RESULTS

Accuracy

Validation based on non-CUP top 1 calls across 90 tumor categories in 3,200 MI Tumor Seek Hybrid cases

Sample Output3 For Cholangiocarcinoma (95%)

TISSUE

  • Adrenal Cortical Carcinoma (score 0.00)
  • Bladder/Urinary Tract (score 0.00)
  • Bowel (score 0.00)
  • Breast (score 0.00)
  • CNS/Brain (score 0.00)
  • Cervix/Uterine Carcinoma (score 0.00)
  • Cutaneous Squamous Cell Carcinoma (score 0.00)
  • Esophagus/Stomach (score 0.00)
  • Germ Cell Tumor (score 0.00)
  • Hematological (score 0.00)
  • Hepatocellular Carcinoma (score 0.03)
  • Kidney (score 0.00)
  • Melanoma (score 0.00)
  • Mesothelioma (score 0.00)
  • Neuroendocrine Neoplasm (score 0.00)
  • Non-Small Cell Lung Carcinoma (score 0.00)
  • Orogenital Squamous Cell Carcinoma (score 0.00)
  • Ovarian Epithelial Tumor (score 0.00)
  • Pancreatobiliary (score 0.97)
  • Peripheral Nervous System (score 0.00)
  • Prostate Adenocarcinoma (score 0.00)
  • Salivary Gland Tumor (score 0.00)
  • Sex Cord Stromal Tumor (score 0.00)
  • Soft Tissue/Bone (score 0.00)
  • Thymic Carcinoma (score 0.00)
  • Thyroid

Pancreatobiliary Expanded

  • Cholangiocarcinoma (score 0.95)
  • Gallbladder Cancer (score 0.00)
  • Pancreatic Adenocarcinoma (score 0.02)

Caris GPSai Tumor Origin Predictor

The Caris GPSai algorithm was trained on Caris genomic and trascriptomic data, covering 90 cancer categories (26 major types and 64 sub-types).

Adrenal Cortical Carcinoma


Bladder/Urinary Tract

  1. Urothelial Carcinoma
  2. Bladder Adenocarcinoma

Bowel

  1. Appendiceal Adenocarcinoma
  2. Colorectal Adenocarcinoma
  3. Small Bowel Carcinoma

Breast

  1. Metaplastic Breast Cancer
  2. Breast Invasive Lobular Carcinoma
  3. Breast Invasive Ductal Carcinoma

CNS/Brain

  1. Diffuse Glioma
  2. Meningioma

Cervix/Uterine Carcinoma

  1. Uterine Serous Carcinoma/Uterine Papillary Serous Carcinoma
  2. Uterine Carcinosarcoma/Uterine Malignant Mixed Mullerian Tumor
  3. Uterine Clear Cell Carcinoma
  4. Cervical Adenocarcinoma
  5. Uterine Endometrioid Carcinoma

Cutaneous Squamous Cell Carcinoma


Esophagus/Stomach

  1. Esophageal Squamous Cell Carcinoma
  2. Esophagogastric Adenocarcinoma
    1. Stomach Adenocarcinoma
    2. Adenocarcinoma of the Gastroesophegeal Junction
    3. Esophageal Adenocarcinoma

Germ Cell Tumor


Hematological


Hepatocellular Carcinoma


Kidney

  1. Renal Cell Carcinoma
    1. Chromophobe Renal Cell Carcinoma
    2. Papillary Renal Cell Carcinoma
    3. Renal Clear Cell Carcinoma
  2. Wilms Tumor

Melanoma


Mesothelioma


Neuroendocrine Neoplasm

  1. Well/Moderately-Differentiated Neuroendocrine Tumor
  2. Poorly-Differentiated Neuroendocrine Carcinoma
    1. Large Cell Neuroendocrine Carcinoma
    2. Small Cell Neuroendocrine Carcinoma
  3. Paraganglioma/Pheochromocytoma
  4. Merkel Cell Carcinoma


Non-Small Cell Lung Carcinoma

  1. Lung Squamous Cell Carcinoma
  2. Lung Adenocarcinoma

Orogenital Squamous Cell Carcinoma


Ovarian Epithelial Tumor

  1. Ovarian Carcinosarcoma/Malignant Mixed Mesodermal Tumor
  2. Serous Ovarian/Fallopian Tube/Peritoneal
    1. Low-Grade Serous Ovarian/Fallopian Tube/Peritoneal Cancer
    2. High-Grade Serous Ovarian/Fallopian Tube/Peritoneal Cancer
  3. Mocinous Ovarian Cancer
  4. Clear Cell Ovarian Cancer
  5. Endometrioid Ovarian Cancer

Pancreatobiliary

  1. Gallbladder Cancer
  2. Pancreatic Adenocarcinoma
  3. Cholangiocarcinoma

Peripheral Nervous System

  1. Schwannoma
  2. Neuroblastoma
  3. Malignant Peripheral Nerve Sheath Tumor

Prostate Adenocarcinoma


Salivary Gland Tumor


Sex Cord Stromal Tumor

  1. Granulosa Cell Tumor
  2. Sertoli-Leydig Cell Tumor

Soft Tissue/Bone

  1. Gastrointestinal Stromal Tumor
  2. Leiomyosarcoma
  3. Angiosarcoma
  4. Rhabdomyosarcoma
  5. Synovial Sarcoma
  6. Osteosarcoma
  7. Liposarcoma
  8. Chondrosarcoma
  9. Endometrial Stromal Sarcoma
  10. Ewing Sarcoma

Thymic Carcinoma


Thyroid

  1. Anaplastic Thyroid Cancer
  2. Well-Differentiated Thyroid Cancer
    1. Papillary Thyroid Cancer
    2. Follicular Thyroid Cancer
  3. Hurthle Cell Thyroid Cancer
  4. Medullary Thyroid Cancer
Caris GPSai can be added to any solid tumor order by selecting the appropriate box on the Caris requisition at no additional cost or added specimen requirements. Results for Caris GPSai will appear in the final Caris report. By comparing the genomic and transcriptomic characteristics of a patient’s tumor to the 90 cancer type signatures, these findings can offer additional insights to make more informed treatment decisions.

Example Caris Report: Caris GPSai (Genomic Probability Score)

  1. Reports a probability score based off both genomic (DNA) and now transcriptomic (RNA) profiling data.
  2. Analyzes genomic and transcriptomic data to match a tumor’s molecular signature to one or more of 90 cancer categories from the Caris database.
  3. Can be added to any solid tumor order at the oncologist’s discretion by selecting the appropriate box on the requisition. The Intent is for the test to only be requested for CUP and cases with atypical clinical presentation or clinical ambiguity.
  4. Must be requested by selecting the appropriate box on the requisition.
  1. Massard C, Loriot Y, Fizazi K. Carcinomas of an unknown primary origin–diagnosis and treatment. Nat Rev Clin Oncol. 2011 Nov 1;8(12):701-10. doi: 10.1038/nrclinonc.2011.158. PMID: 22048624
  2. Peck, M., Moffat, D., Latham, B., Badrick, T., Review of diagnostic error in anatomical pathology and the role and value of second opinions in error preventions, J. Clin. Pathol. 71 (11) (2018) 995-1000.
  3. Graphic generated using OncoTree framework. See Kundra, R., et al (2021) OncoTree: A Cancer Classification System for Precision Oncology. JCO Clin Cancer Inform 5, 221-230

Complete the form below to get in contact with your local representative

Name(Required)

Need Support?
Contact Us
or call 1.888.979.8669 (international: +41 21 533 53 00)