US 12,381,013 B2
Predicting tolerability in aggressive non-hodgkin lymphoma
Tina Geritz Nielsen, Basel (CH); Joseph Nathaniel Paulson, South San Francisco, CA (US); Daniel Jay Schneider, Wynnewood, PA (US); Edward Jean Bataillard, Welwyn Garden City (GB); Carl Wilson Harris, III, South San Francisco, CA (US); Carsten Henneges, South San Francisco, CA (US); and Yoonha Choi, South San Francisco, CA (US)
Assigned to GENENTECH, INC., South San Francisco, CA (US)
Filed by GENENTECH, INC., South San Francisco, CA (US); and HOFFMANN-LA ROCHE INC., Little Falls, NJ (US)
Filed on Jan. 19, 2023, as Appl. No. 18/156,909.
Application 18/156,909 is a continuation of application No. PCT/US2021/070958, filed on Jul. 26, 2021.
Claims priority of provisional application 63/111,777, filed on Nov. 10, 2020.
Claims priority of provisional application 63/060,371, filed on Aug. 3, 2020.
Prior Publication US 2023/0154626 A1, May 18, 2023
Int. Cl. G16H 50/50 (2018.01); G16H 20/10 (2018.01); G16H 50/30 (2018.01)
CPC G16H 50/50 (2018.01) [G16H 20/10 (2018.01); G16H 50/30 (2018.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
accessing an input data set that includes multiple input data values pertaining to a particular subject with lymphoma, each input data value corresponding to a variable of a set of variables, wherein the multiple input data values comprise an index value characterizing comorbidities of the particular subject;
inputting the input data set into a machine-learning model to generate a score corresponding to a degree to which the particular subject will tolerate a particular treatment to the lymphoma, wherein tolerating the particular treatment comprises not having the particular treatment ended or reduced below a threshold dose within a time period after starting the particular treatment, wherein the machine-learning model comprises:
a set of parameters determined using multiple training data elements, each of the multiple training data elements corresponding to a training subject with the lymphoma, and each of the multiple training data elements including a training input data set and a label, the label indicating a tolerance of the training subject to the particular treatment; and
a function relating received input data sets and the parameters to the score;
generating a prediction of the tolerance of the particular subject to the particular treatment using the generated score; and
administering (a) the particular treatment to the particular subject when the prediction of the tolerance is that the particular subject will tolerate the particular treatment, and (b) an alternative treatment to the particular subject when the prediction of the tolerance is that the particular subject will not tolerate or is unlikely to tolerate the particular treatment, wherein the particular treatment is R-CHOP (rituximab, cyclophosphamide, doxorubicin hydrochloride, vincristine, prednisolone), M-CHOP (mosuntuzumab with CHOP), or anti-CD20 treatment with CHOP.