BillionToOne, Inc. has launched Northstar Origin, a liquid-biopsy add-on designed to help oncologists determine the likely tissue of origin when a patient has cancer but conventional pathology and imaging cannot confidently establish where the disease began. The physician-ordered test runs alongside the company’s Northstar Select comprehensive genomic profiling assay and uses the same blood draw, combining cell-free DNA methylation patterns with independent signals from tissue-associated somatic variants. BillionToOne says Northstar Origin identified the correct tissue as its top prediction in 87% of cases in an independent 1,578-patient validation cohort and within its top three predictions in 91%.
The test addresses a relatively small but clinically difficult part of oncology. BillionToOne cites estimates that approximately 3% to 5% of newly diagnosed cancers present as cancer of unknown primary, while uncertain diagnoses may be more common in community settings without access to extensive subspecialty pathology. When the primary site remains unknown, treatment selection becomes more complicated because many oncology regimens, biomarkers and regulatory labels remain tied to the organ or tissue where the cancer originated.
How does Northstar Origin try to identify where a cancer started?
Tumors shed fragments of DNA into circulation, creating circulating cell-free and tumor DNA that can be analyzed from a blood sample. Different tissues carry characteristic patterns of DNA methylation, while particular genomic alterations can also occur with different frequencies across different tumor types. Northstar Origin combines those two information streams rather than relying on one signal alone.
The classifier was developed using 6,025 blood samples from patients with advanced cancer and subsequently evaluated in an independent cohort of 1,578 patients. BillionToOne reported that the correct primary tissue ranked first in 87% of cases and appeared among the top three predictions in 91%. The company plans prospective studies to extend coverage to additional tumor types and evaluate whether the predictions meaningfully improve clinical decision-making.
Why can determining the primary cancer site be so difficult?
Metastatic cancer can occasionally present through lesions in lymph nodes, liver, bone, lung or other organs without an obvious primary tumor. Pathologists can use tissue morphology and immunohistochemistry to infer where the cancer originated, while computed tomography, magnetic resonance imaging, endoscopy and other procedures may search for the primary site. Even after extensive workup, some cases remain uncertain.
That uncertainty has practical consequences because treatment increasingly depends on both tissue type and molecular characteristics. A mutation that supports one therapy in lung cancer may not have the same regulatory or evidence context in another malignancy, while prognosis and expected treatment response can also depend on the primary disease. A credible tissue-of-origin prediction could therefore help oncologists narrow the differential diagnosis and select more focused confirmatory testing.
Does Northstar Origin replace pathology or biopsy?
No. BillionToOne explicitly states that the test is intended to complement, not replace, pathological determination of tissue origin or the broader clinical picture. Northstar Origin can produce false-positive or false-negative results, and its output should be considered alongside imaging, pathology, clinical presentation and other diagnostic information.
This is an especially important limitation because an algorithmic tissue prediction can appear more definitive than it really is. An 87% top-prediction accuracy means the first answer was incorrect in roughly 13% of cases within the validation dataset, and real-world performance can vary depending on cancer type, tumor burden, treatment history and specimen characteristics. The value may therefore lie in guiding physicians toward the right next investigation rather than ending the diagnostic process.
Is Northstar Origin FDA approved?
No. Northstar Origin, Northstar Select and Northstar Response are laboratory-developed tests performed in BillionToOne’s CLIA-certified and CAP-accredited laboratory, and the company states that they have not been cleared or approved by the U.S. Food and Drug Administration.
That distinction is essential when comparing liquid biopsies. Laboratory-developed tests can be clinically offered through certified laboratory frameworks without holding an FDA marketing authorization for the assay. Physicians and patients should therefore interpret BillionToOne’s validation results in the context of the company’s laboratory-developed-test status rather than describing Northstar Origin as an FDA-cleared cancer diagnostic.
What else changed in BillionToOne’s oncology portfolio?
BillionToOne simultaneously expanded Northstar Select from 84 to 102 genes, adding 18 genes and biomarkers intended to reflect recent and emerging therapy-selection needs. The expanded panel includes MTAP deletion, which has become increasingly relevant as drug developers test therapies targeting vulnerabilities associated with MTAP-deficient cancers.
The company previously reported that Northstar Select identified more actionable or pathogenic genomic alterations than aggregated comparators in a prospective 182-patient head-to-head analysis, although comparison methodology and patient mix are important when interpreting manufacturer-reported superiority claims. The strategic logic behind the new launch is to make one blood draw answer two questions: which genomic alterations may guide therapy, and which tissue most likely produced the cancer.
What should oncologists watch next?
Prospective clinical utility is now the key question. An assay can predict tissue origin accurately without necessarily proving that physicians change treatment appropriately or that patients achieve better outcomes because of the result. BillionToOne says prospective studies will evaluate additional tumor types and clinical utility, which should help establish whether the test reduces diagnostic procedures, shortens time to treatment or improves therapy selection.
Northstar Origin nevertheless addresses a genuine diagnostic problem with a practical workflow advantage: no additional tissue and no second blood draw are required when it is ordered with Northstar Select. If future studies confirm that the molecular prediction materially improves decisions, the broader significance could extend beyond cancer of unknown primary and toward a more integrated liquid-biopsy model in which one specimen helps define both what molecular targets a cancer carries and where that cancer most likely began.