Search

EP-4739801-A1 - CLASSIFICATION, TREATMENT, AND TREATMENT RESPONSE OF CANCER PATIENTS

EP4739801A1EP 4739801 A1EP4739801 A1EP 4739801A1EP-4739801-A1

Abstract

The current disclosure describes a method for prediction of a response of a subject with cancer to at least one composition used in the treatment of cancer and/or in the prediction of a cancer disease progression. The method is based on combining information about the overall aggressivity of the cancer and information related to a specific pharmacologic vulnerability of the cancer with other data. The method is particularly useful for breast cancer at the time of first metastasis.

Inventors

  • ANDERSSON, KARL
  • DEMANSE, David
  • HACKL, WOLFGANG

Assignees

  • OncoGenomX

Dates

Publication Date
20260513
Application Date
20240703

Claims (16)

  1. 1. A method for prediction of a response of a subject to at least one compound or composition used in the treatment of metastatic breast cancer of said subject, the method comprising the steps of: subjecting (S1) a first cancer sample of said breast cancer to a first analysis comprising: conducting (S11) measurements of a first set of RNA expression levels and a first set of gene mutation status to obtain a first set of measurement results; obtaining (S12) first information, the first information comprising clinical parameters related to said subject and disease-related parameters related to said first cancer sample; applying (S13) a classification algorithm to said first set of measurement results and said obtained information to estimate overall aggressiveness and assign said first cancer sample to a class of aggressiveness; subjecting (S2) a second cancer sample of said breast cancer to a second analysis comprising: conducting (S21) measurements of a second set of RNA expression levels to obtain a second set of measurement results, said second set of measurement results being indicative of a signature of a pathway related to a specific pharmacologic vulnerability; applying (S22) an algorithm to said second set of measurement results to obtain the signature of said pathway; obtaining (S3) second information, the second information comprising a history of said cancer in said subject, wherein the history of said cancer comprises information regarding a prior adjuvant treatment of said cancer; determining (S4) a pattern at least in part based on a combination of i) the assigned class of aggressiveness of the first cancer sample and ii) the obtained signature of said pathway, wherein said signature of said pathway indicates if said cancer sample has an elevated specific pharmacologic vulnerability; and predicting (S5) a response of the subject to at least one compound or composition based on the pattern.
  2. 2. The method according to claim 1, the method further comprising: providing (S6) a recommendation related to the use of said compound or composition for the treatment of said cancer of said subject based on the pattern and/or predicted response.
  3. 3. The method according to claims 1-2, wherein it is predicted if said compound or composition is compatible as a follow-up treatment to said prior adjuvant treatment.
  4. 4. The method according to claims 1-3, wherein an efficacy of said compound or composition for all aggressiveness classes having the same signature of said pathway in view of said prior adjuvant treatment is known from the history of said cancer comprising historic observations on patterns obtained from at least one reference subject population.
  5. 5. The method according to claim 4, further comprising: classifying (S41) said subject to be treated as being susceptible to treatment with said compound or composition if said historic observations indicate favorable efficacy for said compound or composition for the combination of said cancer sample aggressiveness class and signature of said pathway; or classifying (S42) said subject to be treated as not being susceptible to treatment with said compound or composition if said historic observations lack a favorable efficacy for said compound or composition for the combination of said cancer sample aggressiveness class and signature of said pathway.
  6. 6. The method according to claims 1-5, wherein said breast cancer is hormone receptor positive (HR Pos ) , human epidermal growth hormone receptor negative (HER2 Neg ) breast cancer, and said adjuvant treatment is one of aromatase inhibitor (Al), selective estrogen receptor modulator (SERM), selective estrogen receptor destroyer (SERD).
  7. 7. The method according to claims 1-6, wherein said overall aggressiveness is estimated using the invasive ductal luminal breast cancer (IDLBC) 1-3 classification.
  8. 8. The method according to claims 1-7, wherein said signature of said pathway is one of i) Phosphoinositide 3-kinase (PI3K) signature, ii) Cyclin-dependent kinase 4 (CDK4) signature, iii) DNA damage signature and iv) Drug sensitivity signature.
  9. 9. The method according to claims 2-8, wherein a character of said provided recommendation related to the composition is contradicting said pharmacologic vulnerability suggested by said signature pathway.
  10. 10. The method according to claims 1-8, wherein the second analysis indicates an elevated specific pharmacologic vulnerability of the compound or composition, which vulnerability indicates eligibility of said compound or composition for use in the treatment of metastatic breast cancer of said subject, and wherein said indicated eligibility is considered valid only for selected class(es) of overall aggressiveness, wherein the selected class(es) of overall aggressiveness are selected based on the class of aggressiveness obtained in the first analysis.
  11. 11. The method according to any of the preceding claims, wherein raw data relating to one or more of the first set of measurement results, the first information, the second set of measurement results and the second information is encoded into one or more generalized intermediate entities, wherein each generalized intermediate entity contains latent information of the encoded raw data and is used as input when applying the algorithms to obtain the signature of a pathway related to a specific pharmacologic vulnerability and estimating overall aggressiveness.
  12. 12. The method according to any one of claims 2-9, wherein said signature of said pathway is a Phosphoinositide 3-kinase (PI3K) signature and wherein said provided recommendation related to the use of said composition for the treatment of said cancer comprises recommending a composition comprising at least one of: i) An aromatase inhibitor (Al), ii) A non-steroidal aromatase inhibitor (NSAI), iii) A steroidal aromatase inhibitor (SAI), iv) A selective estrogen receptor modulator (SERM), v) A selective estrogen receptor destroyer (SERD), vi) A Cyclin Dependent Kinase 4 (CDK4) inhibitor, and vii) A PI3K inhibitor.
  13. 13. A method for applying a statistical test on patient data while ensuring patient privacy within a first organization (100) that is allowed to manage patient data, the method comprising: providing (S100) raw data related to a patient suitable for a statistical test; encoding (S110) said raw data into one or more intermediate entities; transferring (S120) said intermediate entities to a third party organization (200) for statistical analysis; upon receiving (S130) a result from said third party organization (200), using (S140) said result to improve care of said patient.
  14. 14. A method for applying a statistical test on patient data while ensuring patient privacy within a third party organization (200) that may or may not be allowed to manage patient data, the method comprising: receiving (S125A) intermediate entities from a first organization (100) that is allowed to manage patient data, the intermediate entities comprising encoded raw data related to a patient suitable for a statistical test; conducting (S125B) a statistical analysis on said intermediate entities; providing (S125C) a result from the statistical analysis to said first organization (100), wherein it is impossible to reverse said intermediate entities into the provided raw data.
  15. 15. The method according to anu one of claims 13 or 14, further comprising: the first organization (100) transferring (S150) a confirmed diagnosis or medical status related to said patient to said third party organization (200); the third party organization (200) improving (S155) said statistical test based on said intermediate entities and said confirmed diagnosis or medical status;
  16. 16. The method according to any one of claims 14-15, further comprising: the third party (200) applying (S165) a statistical test for prediction of a response of a subject to at least one compound or composition used in the treatment of metastatic breast cancer of said subject as defined in any of the claims 1-12.

Description

CLASSIFICATION, TREATMENT, AND TREATMENT RESPONSE OF CANCER PATIENTS TECHNICAL FIELD The present disclosure relates to the selection of anti-cancer treatments for cancer patients. More specifically, the proposed technique relates to selection of one or more treatment based on detailed characterization of the cancer. The disclosure comprises methods for recommending a composition for use in the treatment of cancer, and to the compositions recommended for use in treatment. BACKGROUND Cancer is a major health and economic burden for individuals and society. Response to cancer treatments varies between subjects, tumor subtypes, and over the treatment course. Heterogeneity in cancers is a main cause why individual tumors are not equally sensitive to cancer drugs, even when their tumors are of the same genetic subtype. With a few exceptions, available biomarkers, a number which is constantly increasing, have not changed this situation significantly (Ramon y Cajal et al., J Mol Med 2020, 98(2): 161-177). In Breast Cancer (BC) for example, immunohistochemical receptor analysis (e.g. estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2) gives guidance to physicians to help them determine whether a breast cancer patient is eligible for anti-hormonal treatment. However, receptor measurement does not reliably predict tumor vulnerability to specific anti-hormonal treatment options (e.g. (estrogen) receptor or synthesis blockade), nor treatment outcome. Twenty clinical decision support systems (CDSS) used for prognostication, recurrence/metastasis risk scoring and/or guiding decisions on withholding adjuvant systemic chemotherapy have not changed this situation. Due to the nature of prognostic factors, recurrence/metastasis risk scores and/or the design of these systems they provide a basis for empirical personalized treatment decisions based on clinical outcome trends observed in matching patients with comparable recurrence/metastasis risk. They do, however, not allow for personalized treatment decisions, tailor-made for a patient’s individual cancer, accounting for its tumor biologic properties, its pharmacologic vulnerabilities, the biologic relevance of drug targets for the tumor and/or and its inherent clinical behavior. Available specific cancer gene tests are only poor reflections of the biological escape mechanism the cancer cells of an individual tumor have adopted to effectively grow under the condition of genetic degeneration and increased cellular stress. It is therefore not surprising that molecularly predicted outcome results often deviate from clinical reality (see, e.g., Malone et al., Genome Med. 2020, 12: 8; Harbeck et al., Nature Reviews Disease Primers 2019, 5, 66; Kim et al., Expert Review of Anticancer Therapy, 2020, 20:3, 205-219). Different methods of classification of hormone-dependent breast cancer are known or in use: Classification by expression levels of hormone (estrogen, progesterone) and human epithelial receptor-2 (HER2) as determined by immunohistochemistry. Classification by amplification status of the coding gene of HER2 (ErbB2) as determined by FISH or CISH. Classification by recurrence risk status based on a cancer’s clinical and pathologic disease stage, it’s molecularly determined recurrence risk or mixtures of both. Many attempts to construct a signature of a biological pathway for the purpose of indicating if a particular biological pathway is active have been made. Normally such a signature is made of measurement values of protein concentrations, genotype results or the similar. A signature of a pathway is therefore often a single value which is composed of many underlying input values. A signature of a biological pathway can be associated with a pharmacologic vulnerability, i.e. the sensitivity of a cell to treatment using particular classes of medication. One example of a signature of a biological pathway which is associated with a pharmacologic vulnerability is the PIK3 pathway according to Levine et al (Levine, David M, David R Haynor, John C Castle, Sergey B Stepaniants, Matteo Pellegrini, Mao Mao, and Jason M Johnson. 2006. “Pathway and Gene-Set Activation Measurement from mRNA Expression Data: The Tissue Distribution of Human Pathways.” Genome Biology 7 (10): R93.). Another example is US2011/0307427A1 where an attempt to predict response to adjuvant therapy was made using one signature-like entity. Common features of available tests are their chemotherapy focus, high discordance ranging from 42% to 66% (Varga et al., Int. J. Cancer: 145, 882-893 (2019)) as a reflection of their variably low/to moderated predictive accuracy. Available immunohistochemical tests help with the decision, whether a tumor is eligible for anti-hormonal cancer treatment. Consideration of a patients estimated recurrence/metastasis risk is used to guide extension of anti-hormonal treatment from 5 to 10 years. In addition, attempts have been made to classify a canc