(12) United States Patent (10) Patent No.: US 9,322,066 B2 Ricci Et Al

USOO9322066B2

(12) United States Patent (10) Patent No.: US 9,322,066 B2 Ricci et al. (45) Date of Patent: Apr. 26, 2016

(54) PREDICTORS FOR CANCERTREATMENT FOREIGN PATENT DOCUMENTS (75) Inventors: Deborah Ricci, Ringoes, NJ (US); WO WO 2004/053066 A2 6, 2004 Weimin Li, Radnor, PA (US); Erin WO WO 2006,133420 A2 12/2006 DeVay Henitz, Flemington, NJ (US) WO WO 2008/021183 A2 2, 2008 (73) Assignee: Janssen Pharmaceutica NV, Beerse WO WO 2009,148528 A2 12/2009 (BE) OTHER PUBLICATIONS (*) Notice: Subject to any disclaimer, the term of this Lichter et al., Blood, 2012, 120:4513-4516.* patent is extended or adjusted under 35 Coiffier et al., Blood, ASH Annual Meeting Abstracts 2011; 118: Abstract 265.* U.S.C. 154(b) by 4 days. Ricci et al. Blood, ASH Annual Meeting Abstracts, 2009 114: Abstract 3875. (21) Appl. No.: 13/569,517 Michiels et al, Lancet, 2005, 365:488-492.* de Jong et al. Haematologica, 2009, 94:70-77.* (22) Filed: Aug. 8, 2012 Canioni et al., J. Clin Oncol, 2008, 26:440-446.* International Search Report relating to International Patent Applica (65) Prior Publication Data tion No. PCT/US2012/049941, filed Aug. 8, 2012. Date of Mailing of International Search Report: Jan. 28, 2013. US 2013/0216524 A1 Aug. 22, 2013 Written Opinion of the International Search Authority relating to International Patent Applcation No. PCT/US2012/049941, filed Aug. Related U.S. Application Data 8, 2012. Date of Mailing of Written Opinion: Jan. 28, 2013. Arif et al., “Frequency of bcl-2 gene rearrangement in B-Cell Non (60) Provisional application No. 61/522,596, filed on Aug. Hodgkin's lymphoma.”. Asian Pacific J Cancer Prev, 2009: pp. 237 11, 2011, provisional application No. 61/560,555, 240, vol. 10(2). Binstadt et al., “IgG Fc receptor polymorphisms in human disease: filed on Nov. 16, 2011. Implications for intravenous immunoglobulin therapy.” J Allergy Clin Immunol. 2003, pp. 697-703, vol. 111(4). (51) Int. Cl. Blum, K.A., “Upcoming Diagnostic and Therapeutic Developments GOIN33/574 (2006.01) in Classical Hodgkinls Lymphoma II”, Hematology Dec. 1, 2010, A 6LX39/395 (2006.01) pp. 93-100, vol. 2010(1), XP55049366, ISSN: 1520-4391, DOI: 10. CI2O I/68 (2006.01) 1182, asheducation-2010.1.93, p. 94; table 2. A6 IK3I/96 (2006.01) Catron et al. “Therapeutic activity of humanized anti-CD20 A 6LX3/573 (2006.01) monoclonal antibody and polymorphism in IgGFc receptor Fcgam maRIIIa gene.”. Blood 2002, pp. 757-758, vol. 99(3). A6 IK3I/69 (2006.01) Chen et al., “Autocatalytic Subunit processing couples active site GOIN33/92 (2006.01) formation in the 20S proteasome to completion of assembly”. Cell, A6 IK3I/9 (2006.01) 1996, pp. 961-972, vol. 86. A6 IK3I/98 (2006.01) Chen etal, “Genome-wide siRNA screen for modulators of cell death induced by proteasome inhibitor bortezomib’. Cancer Res., 2010, (52) U.S. Cl. pp. 4318-4326, vol. 70(11). CPC ...... CI2O I/6886 (2013.01); A61 K31/19 Dave et al., “Prediction of Survival in follicular lymphoma based on (2013.01); A61 K31/196 (2013.01); A61 K molecular features of tumor-infiltrating immune cells.”. NEJM 2004, pp. 2159-2169, vol. 351(21). 31/198 (2013.01); A61 K3I/573 (2013.01); Ding et al. “Bortezomib in combination with IGEV chemotherapy A61K3I/69 (2013.01); A61K.39/39558 regimen for a primary refractory Hodgkin's lymphoma of bone.”. (2013.01); G0IN33/57407 (2013.01); G0IN Leukemia Research, Sep. 1, 2009, pp. e170-e172, vol. 33(9), New 33/92 (2013.01); G0IN 2800/52 (2013.01) York, NY, US XPO26222057, ISSN: 0145-2126, 001: 10.1016/J. (58) Field of Classification Search LEUKRES.2009.03.036 retrievedon Apr. 28, 2009Jp. 170-p. e171. None Farinha, P., “Analysis of multiple biomarkers shows that lymphoma associated macrophage (LAM) content is an independent predictor of See application file for complete search history. survival in follicular lymphoma (FL).”. Blood, Sep. 15, 2005, pp. 2169-2174, vol. 106(6), XP55041472., ISSN: 0006-4971. 001:10. (56) References Cited 118.2/blood-2005-04-1565, Abstract. Fischer et al., “New approaches and therapeutics targeting apoptosis U.S. PATENT DOCUMENTS in disease.”. Pharmacol Rev, 2005, pp. 187-215, vol. 57(2). Goyetal, "Potential biomarkers ofbortezomib activity in mantle cell 5,693,617 A 12/1997 Stein et al. lymphoma from the phase 2 PINNACLE trial”. Leukemia & 5,756,764 A 5/1998 Fenteany et al. Lymphoma, 2010, pp. 1269-1277, vol. 51(7). 5,780,454 A 7, 1998 Adams et al. Kamper et al.'Tumor-infiltrating macrophages correlate with 6,018,020 A 1/2000 Attwood et al. adverse prognosis and Epstein-Barr virus status in classical 6,066,730 A 5, 2000 Adams et al. Hodgkin's lymphoma.', Haematologica, Nov. 11, 2010, pp. 269 6,075,150 A 6/2000 Wang et al. 276, vol. 96(2), XP55041465, ISSN: 0390-6078, 001: 10.3324/ 6,083,903. A 7/2000 Adams et al. haematol.2010.031542, Abstract. 6,096,778 A 8/2000 Chatterjee et al. (Continued) 6,297.217 B1 10/2001 Adams et al. 6,465,433 B1 10/2002 Adams et al. Primary Examiner — Mark Halvorson 6,548,668 B2 4/2003 Adams et al. 6,617,317 B1 9, 2003 Adams et al. (57) ABSTRACT 6,713,446 B2 3/2004 Gupta The present invention provides methods of predicting a 6,747,150 B2 6/2004 Adams et al. response to a cancer treatment by determining CD68 level or 6,831,099 B1 12/2004 Crews et al. PSMB1 (P11A) polymorphism in a biological sample and the 6,958,319 B2 10/2005 Gupta presence or quantity of a second biomarker in the patient. The 7,109,323 B2 9, 2006 Plamondon et al. 7,119,080 B2 10/2006 Adams et al. invention also provides kits and methods for treating cancer. 7,422,830 B2 9, 2008 Nam et al. 3 Claims, 8 Drawing Sheets US 9,322,066 B2 Page 2

(56) References Cited 2003, pp. 775-782, vol. 27(9), XP5504 1467, ISSN: 0145-2126, 001:10.1016/SO 145-2126(03)00006-7, Abstract. OTHER PUBLICATIONS Steidl et al., “Macrophages predict treatment outcome in Hodgkin's lymphoma”. Haematologica, Jan. 31, 2011, pp. 186-189, vol. 96(2), Karin et al., “NF-kB in cancer: From innocent bystander to major XP55041473, ISSN: 0390-6078, 001: 10.3324/haematol.2010. culprit”. Nature Rev Cancer, 2002, pp. 301-310, vol. 2(4). 033316, p. 188. Keats et al., “Promiscuous mutations activate the noncanonical NF Steidletal, "Tumor-Associated Macrophages and Survival in Classic kappaB pathway in multiple myeloma.” Cancer Cell, 2007, pp. 131 Hodgkin's Lymphoma.', New England Journal of Medicine, Mar. 11, 144, vol. 12(2). 2010, pp. 875-885, vol. 362(10), XP5504 1466. ISSN: 0028-4793. Lichter, D. I. "Sequence analysis of the 20S proteasome f-subunit 001:10.1056/NEJMoaO905680, Abstract. genes in tumor tissue and cell lines.”. A Thesis in the Field of Taskinen et al., “A high tumor-associated macrophage content pre dicts favorable outcome in follicular lymphoma patients treated with Biotechnology for the Degree of Master of Liberal Arts in Extension rituximab and cyclophosphamide-doxorubicin-vincristine Studies., Nov. 2008, Harvard University. prednisone.” Clin Cancer Res, 2007, pp. 5784-5789, vol. 13(19). Lichter, D. I. "Sequence analysis of the 20S proteasome f-subunit Wojcik, C., “Regulation of apoptosis by the ubiquitin and proteasome genes in tumor tissue and cell lines.”. A Thesis in the Field of pathway”. J Cell Mol Med, 2002, pp. 25-48, vol. 6(1). Biotechnology for the Degree of Master Liberal Arts in Extension Wu et al., “A novel polymorphism of FcgammaRIIIa (CD16) alters Studies., Nov. 2008, Harvard University; Bibliographical informa receptor function and predisposes to autoimmune disease.”. J. Clin tion, downloaded from internet Dec. 16, 2013. Invest, 1997, pp. 1059-1070, vol. 100(5). Lichter et al. Supplemental Material, 2012. Yeung et al., “PS-341 (bortezomib) induces lysosomal cathepsin b Mitsiades et al., “Molecular sequelae of proteasome inhibition in release and a caspase-2 dependent mitochondrial permeabilization human multiple myeloma cells.”. Proc Natl Acad Sci, 2002, pp. and apoptosis in human pancreatic cancer cells.”. J. Biol Chem, 2006, 14374-14379, vol. 99(22). pp. 11923-11932, vol. 281 (17). Mulligan et al., “Gene expression profiling and correlation with clini Yin et al., “Proteasome inhibitor PS-341 causes cell growth arrestand cal outcome in clinical trials of the proteasome inhibitor apoptosis in human gliobastoma multiforme (GBM).”. Oncogene, bortezomib”. Blood, 2007, pp. 3177-3188, vol. 109(8). 2005, pp. 344-354, vol. 24(3): 344-354. San Miguel et al. “Bortezomib plus Melphalan and Prednisone for Barton, Mary Kay, “Predictive biomarkers may help individualize Initial Treatment of Multiple Myeloma”. N. Engl. J. Med., Aug. 28. treatment for patients with follicular lymphoma', CA: A Cancer 2008, pp. 906-917, vol. 359(9). Journal for Clinicians, vol. 63, No. 5, Jul. 10, 2013, pp. 293-294. Schmidt et al. “Sequence information within proteasomal XPO55174933, ISSN: 0007-9235, DOI: 10.3322/caac.21197. prosequences mediates efficient integration of 3-subunits into the Coiffier, B., et al., “Prespecified Candidate Biomarkers Identify Fol 20S proteasome complex.”. J Mol Biol. 1999, pp. 117-128, vol. licular Lymphoma Patients Who Achieved Longer Progression-Free 288(1). Survival with Bortezomib-Rituximab Versus Rituximab’. Clinical Shin et al. “Notch1 augments NF-kappaB activity by facilitating its Cancer Research, vol. 19, No. 9, Apr. 2, 2013, pp. 2551-2561, nuclear retention.”. EMBO, 2006, pp. 129-138, vol. 25(1). XP055174935, ISSN: 1078-0432, DOI: 10.1158/1078-0432. CCR Steensma et al. “Splenic histopathological patterns in chronic 12-3069. myelomonocytic leukemia with clinical correlations: reinforcement of the heterogeneity of the syndrome.”. Leukemia Research, Sep. 1, * cited by examiner U.S. Patent Apr. 26, 2016 Sheet 1 of 8 US 9,322,066 B2

Prior lines of treatment =

Excluded CD68 Pily ollicular FLIP (Low) Excluded OR PSMB 1/PI A(C/G) ES No

CD68 Positive Follicular (0-50) ES Selected

Figure 1. U.S. Patent Apr. 26, 2016 Sheet 2 of 8 US 9,322,066 B2

Prior lines of treatment

-H

CD68 Positive"E) Follicul OCU Excluded

Figure 2. U.S. Patent Apr. 26, 2016 Sheet 3 of 8 US 9,322,066 B2

Prior lines of treatment - 1 Excluded

Selected

Figure 3. U.S. Patent Apr. 26, 2016 Sheet 4 of 8 US 9,322,066 B2

CD68 Overal OR Positive Follicular Low Tumor Burden (0-50)

CD68 Selected Positive Follicular (0-50) Excluded Excluded

Figure 4. U.S. Patent Apr. 26, 2016 Sheet 5 of 8 US 9,322,066 B2

CD68 Positive Follicular (O-50)

Excluded

Figure 5. U.S. Patent Apr. 26, 2016 Sheet 6 of 8 US 9,322,066 B2

P65 cytoplasmic signals.1

Selected OR

CD68 Positive Follicular (0-50) Excluded

CD68 Overal OR Positive Follicular (0-50) | Selected

Selected

Excluded Excluded

Figure 6. U.S. Patent Apr. 26, 2016 Sheet 7 of 8 US 9,322,066 B2

P65 cytoplasmic Selected CD68 Positive Follicular (0-50) signal al+ & PSMB5/R24CC/T Excluded

Low Tumor Burden

PSMB 1/P11AOR (CFG selectedSelected OR

Excluded Figure 7. U.S. Patent Apr. 26, 2016 Sheet 8 of 8 US 9,322,066 B2

PSMB 1/P11A(C/G)

Selected

Excluded Selected

Figure 8. US 9,322,066 B2 1. 2 PREDICTORS FOR CANCERTREATMENT outcome in response to treatment. "Selected” means that the patient will have an increased chance for favorable outcome CROSS REFERENCE TO RELATED in response to treatment. APPLICATIONS 5 DETAILED DESCRIPTION OF THE INVENTION This application claims priority to provisional application Ser. No. 61/522,596, filed Aug. 11, 2011 and provisional The present invention describes predictors that serve as application Ser. No. 61/560,555, filed Nov. 16, 2011, both of useful tools for the prognosis and planning for the treatment which are incorporated herein by reference in their entirety. of cancer. The predictors are predictive of whether there will 10 be a favorable outcome in response to a particular treatment, FIELD OF THE INVENTION for example, treatment with a proteasome inhibitor. Without limitation, the present invention provides (a) The invention is related to treatment of cancer patients. methods for a predicting response to a treatment in a cancer BACKGROUND OF THE INVENTION patient by determining presence or quantity of one or more 15 predictors, (b) kits useful in determination of the presence or While advances in development of successful cancerthera quantity of the one or more predictors, (c) methods for treat pies progress, only a Subset of patients respond to any par ing cancer by selecting patients based on presence or quantity ticular therapy. With the narrow therapeutic index and the of one or more predictors and (d) treating cancer in patients toxic potential of many available cancer therapies, such dif with one or more predictors. ferential responses potentially contribute to patients under In certain embodiments, a method is provided for predict going unnecessary, ineffective and even potentially harmful ing response to a cancer treatment (for example, treatment therapy regimens. with a proteasome inhibitor Such as bortezomib) in a cancer One way to optimize therapy to treat individual patients is patient comprising determining the presence or quantity of a to determine whether the patient one or more predictors that predictor in a patient or a biological sample from the patient; correlate with a particular outcome in response to therapy. 25 and wherein the presence or quantity of the predictor is cor See, e.g., WO2004/053066: WO2006/133420; WO2008/ 021 183; and WO2009/148528. The ability to predict drug related with at least one positive outcome. Certain embodi sensitivity in patients is particularly challenging because drug ments comprise determining the presence or quantity of a responses reflect both the properties intrinsic to the target second predictor in the patient or a biological sample from the cells and also a host's metabolic properties. patient, wherein the presence or quantity of the second pre There is a need to identify further predictive markers to 30 dictor is correlated with at least one positive outcome. identify particular cancer patients who are expected to have a The present invention involves the identification of predic favorable outcome when administered particular cancer tors also referred to herein as “variants”, “markers’ “biomar therapies. kers' and/or “factors', that correlate with an increased prob ability of favorable response to a cancer treatment. The SUMMARY OF THE INVENTION 35 association of patient response to a cancer treatment with these predictors can increase of higher confidence in the The invention provides a method for identifying whethera safety and/or efficacy with the particular treatment. The pre patient has an increased chance for a favorable outcome in dictors may be a gene, protein, patient characteristic, or response to a cancer treatment, comprising: determining the aspect of the patient history. presence, absence or quantity of one or more predictors in the 40 Predictors according to this invention which correlate with patient, wherein the presence, absence or quantity of the at least one favorable outcome include low CD68, PSMB1 predictor correlates with at least one favorable outcome. (P11A) polymorphism, PSMB5 (R24C) polymorphism, P65, The presence of predictors may be determined by obtain age (under 65), one prior treatment, low Follicular Lym ing a biological sample from said patient. The cancer treat phoma International Prognostic Index (FLIPI) score, time ment may comprise administration of a proteasome inhibitor, 45 since last anti-cancer treatment and low tumor burden. Pref such as bortezomib. The predictors may be one or more of low erably, the patient has low CD68 or PSMB1 (P11A) and the CD68, PSMB1 (P11A), PSMB5 (R24C), P65, time since last presence of at least one other predictor. In one embodiment, cancer treatment, one prior treatment, low FLIPI score, age the patient has low CD68 and PSMB1 (P11A)polymorphism. (65 or younger), and low tumor burden. Predictor pairs according to this invention which correlate Also provided are diagnostic kits for identifying patients 50 with at least one favorable outcome include those shown in likely to have a positive outcome in response to a cancer Tables 6 and 7. treatment. By “low CD68 is meant that the subject or biological The invention also provides methods for treating cancer sample from the patient shows less CD68 quantity than the patients by determining the presence, absence or quantity of average patient or biological samples from an average patient one or more predictors in the patient, and selecting a method 55 who has the same disease. In certain embodiments, low CD68 of treatment dependent on whether the patient is likely to means that 25% or less of the cells in a biological sample respond to the treatment. express CD68; 50% or less of the cells in a biological sample Also provided are uses for proteasome inhibitors for the express CD68; 25% or less of the follicular cells in a biologi treatment of cancer, wherein the patients are characterized by cal sample express CD68; 50% or less of the follicular cells in one or more of: low CD68, PSMB1 (P11A), PSMB5 (R24C), 60 a biological sample express CD68; 25% or less of the peri P65, time since last cancer treatment, one prior treatment, low follicular cells in a biological sample express CD68; or 50% FLIPI score, age, and low tumor burden. or less of the perifollicular cells in a biological sample express CD68. DESCRIPTION OF THE FIGURES By “low FLIPI score is meanta score of 0 or 1 factor on the 65 Follicular Lymphoma International Prognostic Index (FLIPI FIGS. 1-8 show decision trees for determining whether a score). To determine FLIPI score, one point is assigned to particular patient will have an increased chance for favorable each of age greater than 60 years, Stage III or IV disease, US 9,322,066 B2 3 4 greater than 4 lymph node groups involved, serum hemoglo Vincristine, vinblastine, Vindesine, and Vinorelbine. In a pre bin less than 12 g/dL and elevated serum LDH. ferred embodiment, the cancer treatment comprises ritux As used herein, the terms "comprising”, “containing. imab. In other preferred embodiments, the cancer treatment "having and “including are used in their open, non-limiting comprises melphalin or prednisone, or a combination of mel SSC. phalin and prednisone. "Quantity” may mean the value, intensity, concentration, In certain embodiments, the cancer treatment is a combi amount, degree, or expression level. For example, quantity of nation treatment. The combination treatment may comprise a gene may be the number of times a gene or portion thereof treatment with a proteasome inhibitor and another cancer is present in a Subject's genome or in the cells of the Subject. treatment or anti-cancer agent. In certain embodiments, the Quantity may also mean the number of cells in a biological 10 sample expressing a marker, or the overall expression level or other anti-cancer agent is a monoclonal antibody, e.g., ritux intensity of the marker in a biological sample. Quantity may imab. In other embodiments, the other anti-cancer agent is also refer to the number of types or lines of therapy the patient melphalin, prednisone, or a combination of melphalin and to which the patient may previously been exposed. The quan prednisone. tity may be in comparison to an absolute number, in compari 15 The favorable outcome may be an overall response rate, son to a reference sample from a healthy patient, in compari overall Survival rate, overall complete response rate, duration son to an average number from healthy patients, or in of response, longer time to next therapy, treatment free inter comparison to an average number from patients with similar val, positive response to treatment, a longer time-to-progres disease. Sion, longer term Survival and/or longer progression-free Sur The cancer treatment may include administration of a vival. The favorable outcome may be dose-dependent or single drug or treatment, or a combination treatment compris dose-independent. The favorable outcome may favorable be ing administration of more than one drug or treatment. The in comparison to no treatment, or in comparison to another cancer treatment may be administration of chemotherapy, cancer treatment or cancer treatment(s). radiotherapy, or immunotherapy, or the cancer treatment may “Cancer or “tumor is intended to include any neoplastic be a bone marrow transplant. 25 growth in a patient, including an initial tumor and any In certain embodiments, the cancer treatment comprises metastases. The cancer can be of the hematological or Solid administering a proteasome inhibitor to a patient. A protea tumor type. Hematologic cancers include Such as myelomas Some inhibitor is any Substance which inhibits enzymatic e.g., multiple myeloma), leukemias (e.g., Waldenstrom's activity of the 20S or 26S proteasome in vitro or in vivo. In syndrome, acute myelogenous leukemia, chronic lympho Some embodiments, the proteasome inhibitor is a peptidyl 30 cytic leukemia, granulocytic leukemia, monocytic leukemia, boronic acid. Peptidylboronic acids include bortezomib. Pro teasome inhibitors include those compounds disclosed in lymphocytic leukemia), and lymphomas (e.g., follicular lym U.S. Pat. Nos. 5,756,764; 5,693,617; 6,831,099; 6,096,778: phoma, mantle cell lymphoma, diffuse large B cell lym 6,075,150; 6,018,020; 7,119,080; 6,747,150; 6,617,317; phoma, malignantlymphoma, plasmocytoma, reticulum cell 6,548,668; 6,465,433; 6,297,217; 6,083,903; 6,066,730; 35 sarcoma, Hodgkin’s disease, non-Hodgkin’s lymphoma or 5,780,454; 7,422,830; 7,109,323; 6,958,319; 6,713,446; and follicular B-cell non-Hodgkin’s lymphoma). Solid tumors 6,699,835. The proteasome inhibitor may be bortezomib. can originate in organs, and include cancers such as brain, In certain embodiments, the cancer treatment comprises skin, lung, breast, prostate, ovary, colon, kidney, and liver. treatment with anti-cancer agents, including but not limited The cancer may be at the primary site, a metastasis, refractory to, acemannan, aclarubicin, aldesleukin, alemtuzumab, ali 40 (e.g. refractory to one or more lines of treatment) and/or tretinoin, altretamine, amifostine, aminoglutethimide, amsa recurring. In certain embodiments, the cancer is follicular crine, anagrelide, anastroZole, ancestim, asparaginase, beva B-cell non-Hodgkin’s lymphoma or multiple myeloma. cizumab, bexarotene, broXuridine, capecitabine, When the predictor is present within the patient’s body, the celmoleukin, cetrorelix, cetuximab, cladribine, clofarabine, presence, absence or quantity of the predictor may be clotrimazole, daclizumab, dexraZOxane, dilaZep, docosanol, 45 assessed by obtaining a biological sample from a patient and doxifluridine, bromocriptine, carmustine, cyclophospha determining whether said biological sample contains the pre mide, cytarabine, diclofenac, edelfosine, edrecolomab, eflo dictor or in what amounts the biological sample contains the rnithine, emitefur, exemestane, exisulind, fadrozole, predictor. A “biological sample' as used herein refers to a filgrastim, finasteride, fludarabine phosphate, formestane, sample containing or consisting of tissues, cells, biological fotemustine, gallium nitrate, gemcitabine, glycopine, hepta 50 fluids and isolates thereof, isolated from a subject, as well as platin, hydroxyurea, ibandronic acid, imiquimod, ioben tissues, cells and fluids present within a Subject. Examples of guane, irinotecan, irsogladine, lanreotide, leflunomide, biological samples include, for example, sputum, blood, lenograstim, lentinan Sulfate, letrozole, liarozole, lobaplatin, blood cells (e.g., white blood cells), amniotic fluid, plasma, lonidamine, masoprocol, melarsoprol, melphalan, mercap serum, semen, saliva, bone marrow, tissue or fine-needle topurine, methotrexate, metoclopramide, mifepristone, 55 biopsy samples, urine, peritoneal fluid, pleural fluid, and cell miltefosine, mirimoStim, mitoguaZone, mitolactol, mitomy cultures. Biological samples may also include sections of cin, mitoxantrone, molgramoStim, nafarelin, nartograstim, tissues such as frozen sections taken for histological pur nedaplatin, nilutamide, noscapine, oprelvekin, osaterone, poses. In certain embodiments, the biological sample may be Oxaliplatin, pamidronic acid, pegaspargase, pentosan or include tumor cells. Biological samples from a hemato polysulfate Sodium, pentostatin, picibanil, pirarubicin, por 60 logical tumor may include bone marrow and/or peripheral fimer Sodium, prednisone, raloxifene, raltitrexed, rasbu blood. ricase, rituximab, romurtide, SargramoStim, sizofuran, Sobu Detection of predictor in a biological sample may be per Zoxane, Sonermin, steroids, Suramin, tasonermin, tazarotene, formed by any conventional method for detecting the type of tegafur, temoporfin, temozolomide, teniposide, tetrachloro predictor, e.g., direct measurement, immunohistochemistry, decaoxide, thalidomide, thymalfasin, thyrotropin alfa, topo 65 immunoblotting, immunoflourescense, immunoabsorbence, tecan, toremifene, trastuzumab, treosulfan, tretinoin, trilos immunoprecipitations, protein array, flourescence in situ tane, trimetrexate, ubenimex, valrubicin, verteporfin, hybridization, FACS analysis, hybridization, in situ hybrid US 9,322,066 B2 5 6 ization, Northern blots, Southern blots, Western blots, cal behavior, morphologic appearance, immunologic, and ELISA, radioimmunoassay, gene array/chip, PCR, RT-PCR, molecular phenotype. Follicular lymphoma (FL), the most or cytogenetic analysis. common indolent NHL, exhibits similar variability with When the predictor is based on a particular genotype or Some patients exhibiting very slow disease course while oth polymorphism, the biological sample may be analyzed by ers progress and die within only 5 year (Dave 2004). A ran genotyping. The term 'genotype' refers to the alleles present domized, open-label, active-controlled, multicenter, multina in DNA from a subject or patient, where an allele can be tional, prospective study to compare the efficacy and safety of defined by the particular nucleotide(s) present in a nucleic acid sequence at a particular site(s). Often a genotype is the the combination of bortezomib and rituximab (Vc-R) to nucleotide(s) present at a single polymorphic site known to single-agent rituximab in Subjects who have relapsed or vary in the human population. “Genotyping refers to the 10 refractory, rituximab-naive or -sensitive follicular B-cell process of determining the genotype of an individual by the NHL was performed. use of biological assays. Current methods of doing this Subjects were centrally randomized to Treatment Groups include PCR, DNA sequencing, antisense oligonucleotide A (Vc-R) or B (rituximab) in a 1:1 ratio taking into account probes, and hybridization to DNA microarrays or beads. the following stratification factors: A “single nucleotide polymorphism” (SNP pronounced 15 Follicular Lymphoma International Prognostic Index Snip) is a DNA sequence variation occurring when a single (FLIPI) score (low 0 or 1 factor, intermediate 2 fac nucleotide-A, T, C, or G—in the genome (or other shared tors, high e3 factors); sequence) differs between members of a species (or between Prior rituximab therapy (yes, no); paired chromosomes in an individual). For example, two Time since last dose of anti-lymphoma therapy (s.1 year, sequenced DNA fragments from different individuals, >1 year); AAGCCTA to AAGCTTA, contain a difference in a single Region (United States/Canada, European Union, Rest of nucleotide. In this case it is said that there are two alleles: C World). and T. Almost all common SNPs have only two alleles. Tumor samples for DNA and protein analysis and a blood The detection of the presence or absence of at least one sample for DNA analysis were collected. genotype variance involves contacting a nucleic acid 25 Protein candidates selected were NF-kB (RELA?p65), sequence corresponding to one of the genes identified herein PSMA5, p27, and CD68. These proteins are attenuated by or a product of such a gene with a probe. The probe is able to bortezomib treatment (NF-kB (RELA?p65); PSMA5, p27), distinguish a particular form of the gene or gene product or regulated by the ubiquitin proteasome pathway (p27), or the presence or a particular variance or variances, e.g., by differential binding or hybridization. associated with poor prognosis in lymphoma (CD68). When the predictor is the presence or quantity (including 30 Elevated expression levels of NF-kB (RELA/p65) were asso the expression level) of a particular gene or protein, the pres ciated with longer time-to-progression (TTP) in mantle cell ence or quantity (including the expression level) may be lymphoma (MCL) and in multiple myeloma (Goy 2010, Mul determined by immunohistochemistry of a biological sample. ligan 2007, and Keats 2007). Low level expression of PSMA5 In certain embodiments, a kit is provided for identifying was associated with longer TTP in MCL (Goy 2010). Sur patients who are candidates for a cancer treatment comprising 35 vival analysis in the MCL study also showed that high levels a first reagent for detecting the presence or quantity of one of of p27 correlated with better overall survival (OS) (Goy the predictors of the invention in a biological sample and a 2010). CD68 was also prespecified and has recently been second reagent for detecting the presence or quantity of a reported to be a prognostic marker for poor outcome in lym second predictor of the invention in a biological sample, and phoma and is also associated with response to rituximab. instructions for employing the predictors to identify patients 40 Candidate genes selected for somatic mutation analysis were who are candidates for the treatment. In certain embodiments, Bcl-2 and Notch-1. Other candidates were considered, but the first reagent detects CD68 quantity and the second reagent were not included because the frequency of the known muta detects PSMB1 (P11A) polymorphism or PSMB5 (R24C) tions was less then 10% in the lymphoma setting. Addition polymorphism. The reagents may be antibodies (for example, ally, small amounts of DNA were recovered from collected when testing CD68) or they may be probes or arrays of probes 45 samples, so the analyses were limited to these two candidates. (for example, when detecting gene polymorphism) Bcl-2 has mutation frequencies of 23% in B-cell lymphoma In certain embodiments, a method for treating a patient for cancer comprising: determining the presence or quantity of a and 50% in follicular lymphoma (Arif 2009) and Notch-1 has first predictor in patient or a biological sample from said a mutation frequency of 24%. Bcl-2 is an important anti patient; and determining the presence or quantity of a second apoptotic protein frequently over expressed in aggressive predictor in said patient or a biological sample from said 50 lymphomas and previous reports have suggested that bort patient; and selecting a method of treatment dependent on eZomib overcomes Bcl-2-mediated protection (Fischer 2005, whether said patient is likely to respond to said treatment. Yin 2005, Yeung 2006, Mitsiades 2002 and Wojcik 2002). The invention also provides uses of proteasome inhibitors Notch-1 has been shown to increase the residence time of for the treatment of cancer in a patient, where the patient is NF-kB (RELA/p65) in the nucleus (Shin 2006). This acts in characterized by the presence, absence, or quantity of at least 55 direct opposition to bortezomib, which prevents NF-kB from one predictor correlated with at least one positive outcome in reaching the nucleus by inhibiting proteasomal degradation response to the proteasome inhibitor. of I-kappa-B proteins, whose role is to retain NF-kB in the All publications cited herein are hereby incorporated by cytoplasm. Notch 1-mediated increased residence time of reference. Unless defined otherwise, all technical and scien NF-kB in the nucleus activates the transcription of the cell tific terms used herein have the same meaning as commonly 60 cycle regulators, eg, cyclins D1 and D2, which may contrib understood to one of ordinary skill in the art to which this ute to the up-regulation of genes involved in immune and invention pertains. inflammatory processes (Karin 2002). Mutations found in functional sequences of these two genes were hypothesized to EXAMPLE1 contribute to inter-individual responses to treatment with 65 bortezomib. Non-Hodgkin lymphoma (NHL) encompasses several Drug target candidate genes were included for both bort unique malignantlymphoid disease entities that vary inclini ezomib (PSMB1, 2, 5, 6, 8, 9) and rituximab (FCGR2A, US 9,322,066 B2 7 FCGR3A). The chemical structure of bortezomib interacts PSMB8 (G8R) with PSMB subunits 1, 2, and 5 and there is documented PSMB9 (R60H, V32I) evidence of polymorphisms in these subunits as well as in PSMA (positive nuclear and cytoplasmic staining) PSMB6, 7, and 8. Polymorphisms within the subunits may CD68 (overall positive, positive follicular, positive peri-fol affect the ability of the drug to bind effectively or may prevent licular) autocatalytic processing of pro-sequences that could lead to P27 (nuclei positive, intensity score) variability in levels of proteasomes and/or response to bort RELA/p65 (positive nuclear and cytoplasmic staining, inten eZomib in individual patients. For rituximab, the presence of sity score). single nucleotide polymorphisms (SNPs) corresponding to The intent-to-treat (ITT) population was defined as all phenotypic expression of valine (V) or phenylalanine (F) at 10 Subjects who were randomized. Subjects in this population amino acid 158 of FCGR3A and of histidine (H) or arginine were analyzed according to the treatment to which they were (A) at amino acid 131 of FCG2A greatly influences the affin randomized. Biomarker evaluations were done on this popu ity of IgG for the Fcy receptor (Binstadt 2003, Wu 1997). lation when biomarker data was generated for the clinical Expression of the high-affinity V allele at 158 results in study endpoints. tighter binding of FCG3A to IgG1 and IgG3, whereas the 15 Protein marker expression levels (CD68, NF-kB (P65), low-affinity Fallele is associated with decreased binding of PSMA5, P27) were determined by immunohistochemistry FCG3A to IgG. Similarly, the high-affinity Hallele at 131 (IHC). The expression level cut-points for single-marker results in greater affinity of FCGR2A for IgG2, whereas the analyses were: low-affinity A allele correlates with decreased binding. Cor CD68: relation of these low-affinity polymorphisms has been asso % positive follicular (0-25, 26-50, 51-75, >75), ciated with worse clinical response and progression-free Sur % positive peri-follicular (0-25, 26-50, 51-75, >75), vival (PFS) after rituximab therapy in studies of NHL overall% positive (0-25, 26-50, 51-75, >75) (Cartron 2002). Therefore, we examined whether subjects NF-kB (p65): with these polymorphisms were responsive to treatment with % positive cytoplasmic signal (<90%, >91% cutoffs), Vc-R as this may be an alternative therapy for patients with 25 % positive nuclear signal (0. <5%, >5%) limited treatment options. Nuclear staining intensity (<1+, >2+) The exploratory objectives in this study were to identify PSMAS: patient populations who were more or less likely to respond to % positive cytoplasmic signal (0-20, 30-50, 60-70, 80-90) Vc-R or rituximab alone by: % positive nuclear signal (0-20, 30-50, 60-70, 80-90) Performing association analyses of CD68, NF-kB (RELA/ 30 Cytoplasmic staining intensity (<2+, 3+) p65), PSMA5, and p27 protein expression levels with Nuclear positive vs. all other selected clinical study endpoints P27: Performing association analyses of Notch-1 and Bcl-2 % positive nuclear staining (0-20, 30-50, 60-70, 80-100) Somatic mutations (single and in combination) with Nuclear signal intensity (<1+, >2+) Selected clinical study endpoints 35 Cut-points selected for pair-wise comparisons were chosen Performing association analyses of FCGR2A and to reduce the total number of comparisons that would be FCGR3A polymorphisms (SNPs) with selected clinical done. The selected cut-points are found in Table 1: study endpoints PSMB1, PSMB2, PSMB5, PSMB6, PSMB8, and PSMB9 TABLE 1 polymorphisms (SNPs) with selected clinical study end 40 points Cut-Points for Protein Markers Included Performing combinations of biomarkers with selected in the Pair-Wise Comparisons clinical study endpoints. Protein Marker Cutoff Multiple testing corrections were done using the false dis 20S (PSMA5)% nuclear staining s20 vs. 20 covery rate (FDR) method for pair-wise comparisons and by 45 20S (PSMA5)% positive cytoplasmic signal s90 wS. c-90 forward selection when multiple biomarker combinations 2OS intensity cytoplasmic signal s2+ vs. 2 were compared. In practical terms, the FDR is the expected CD68 overall positive ssO wS. -SO proportion of false positives; for example, if 1000 observa CD68 positive follicular ssO wS. -SO CD68 positive perifollicular ssO wS. -SO tions were predicted to be different, and the FDR for these P27% nuclei positive sf() ws. >70 observations was 0.10, then 100 of these observations would 50 P27 signal intensity s1+ vs. -1 be expected to be false positives. Over-fitting is controlled by P65 (NF-kB) % nuclear staining Ows. -O P65 (NF-kB) % positive cytoplasmic signal s90% wS. c-90% cross validation and independent validation within the analy P65 (NF-kB) intensity cytoplasmic signal s1+ vs. -1 S1S. Genes with sufficient variation include: FCGR2A (H166R, Q62R, Q62X) 55 Germline SNP data for PSMB Subunits and FCGR2A and FCGR3A (V212F) FCGR3A genes were generated by standard polymerase PSMB1 (P11A) chain reaction (PCR) methodologies. Alleles detected in PSMB5 (R24C) these assays are found in Table 2: TABLE 2

Alleles for PSMB Subunits and FCGR2A and FCGR3A Genes

PSMB1 PSMB2 PSMB5 PSMB6 PSMB8 PSMB9

H166R (aka H131R) US 9,322,066 B2 9 10 TABLE 2-continued

Alleles for PSMB Subunits and FCGR2A and FCGR3A Genes

FCGR2A FCGR3A PSMB1 PSMB2 PSMB5 PSMB6 PSMB8 PSMB9

1208N P107A P11A P193L

S231A

(aka V158F)

The following clinical endpoints were included in the Islander, and Black/Other), FLIPI score (low 0 or 1 factor. analysis within each treatment group and an overall compari intermediate 2 factors, high e3 factors); prior rituximab son was made with the biomarker-related endpoints: Progres therapy (yes, no); time since last dose of anti-lymphoma sion-free survival, defined as the interval between the date of therapy (s.1 year, >1 year); Ann Arbor stage (I, II, III, IV), randomization and the date of progressive disease (PD) or number of prior lines of therapy (1, 2 and more), high tumor death, whichever is first reported in the ITT population; over burden (yes, no), and region (United States/Canada, Euro all survival; overall response rate (complete response CR+ pean Union, Rest of World) and were compared with sum CR unconfirmed ICRu-partial response PRI); overall CR mary statistics from the clinical trial data sets. rate (CR+CRu); duration of response; time to next anti-lym The stratified log rank test and Cox proportional hazard phoma therapy; and treatment free interval. model was utilized for assessments of PFS between the treat Subjects from who met all of the following criteria were ment groups. The Kaplan-Meier method was to be used to included in the analysis: subjects in the ITT population; sub estimate the distribution of overall PFS for each treatment jects with evaluable biomarker data determined by IHC- or 30 group in the biomarker population, and overall; Stratified by PCR-based methodologies; and subjects with clinical data for expression level, SNP, or mutation (or covariate groups as at least one of the clinical endpoints listed above. noted above). The hazard ratio and 95% confidence interval The primary biomarker analysis was aimed at identifica was based on a stratified Cox's proportion hazard model with tion of differentially expressed proteins, mutations, or geno treatment as the explanatory variable. Analyses were done for types that were associated with clinical study endpoints. 35 the population overall and by treatment group, and each of Covariates included in the analysis were: FLIPI score (low 0 these analyses were also stratified by the following factors if or 1 factor, intermediate 2 factors, high e3 factors); prior Sufficient sample size existed: rituximab therapy (yes, no); time since last dose of anti FLIPI score (low 0 or 1 factor, intermediate 2 factors, high lymphoma therapy (s.1 year, >1 year); region (United States/ >3 factors) Canada, European Union, Rest of World), age, sex, race, Ann 40 Prior rituximab therapy (yes, no) Arbor stage (I, II, III, IV), Number of prior lines of therapy (1. Time since last dose of anti-lymphoma therapy (s.1 year, >1 2 and more), and high tumor burden (yes, no). year) For single-marker association analyses and pair-wise com Region (United States/Canada, European Union (Belgium, parisons, a log rank test and Cox proportional hazard model Czech Republic, Finland, France, Germany, Great Britain, was utilized for assessments of PFS, TTP, and OS between the 45 Greece, Hungary, Italy, Poland, Portugal, Slovakia, Spain, treatment groups. Kaplan-Meier curves were utilized to esti and Sweden), Rest of World (Argentina, Australia, Brazil, mate the distribution differences between groups in the time India, Israel, Mexico, China, Korea, Romania, Russia, South to-event analyses. Comparison of the response rates between Africa, Thailand, and Ukraine)) the treatment groups was conducted using the Fishers exact Age (<65, >65) test. Single-marker association analyses were stratified by the 50 Sex covariates. Race For pair-wise comparison analysis, biomarker pairs were Ann Arbor stage (I, II, III, IV) formed by combinations of two markers. A log rank test was Number of prior lines of therapy (1, 2 and more) utilized for assessments of PFS, TTP, and OS between the High tumor burden (yes, no) treatment groups in the Subpopulation defined by biomarker 55 Overall Survival was measured from the date of random pairs. Kaplan-Meier curves were utilized to estimate the dis ization to the date of the subject’s death. If the subject was tribution differences between groups in the time-to-event alive or the vital status was unknown, it was censored at the analyses. Comparison of the response rates between the treat date that the subject was last known to be alive. Similar to PFS ment groups was conducted using the Fishers exact test. and TTP, the Cox proportional hazard model was used to Methods of Analysis for the multiple biomarker comparison 60 evaluate the association between biomarker endpoints and models can be found in section 4.6. OS. Kaplan-Meier survival curves were presented. Analyses For each analysis, demographic and baseline characteristic were done for the population overall and by treatment group, variables were to be summarized for the biomarker popula and each of these analyses was stratified by the covariates tion as follows. Descriptive statistics (mean, standard devia used for PFS. tion, median, and range) were calculated for baseline demo 65 Comparison of the response rates between the treatment graphic data (including: age (year), age category (>65 years groups was conducted using the Fishers Exact Test. Duration ands.65 years), sex (male, female), race (White, Asian/Pacific of response, time to response, and time to clinical relapse US 9,322,066 B2 11 12 were analyzed descriptively, where appropriate, and the more), high tumor burden (yes, no), region (United States/ biomarker Subset was compared (if appropriate, determined Canada, European Union, Rest of World), sex, and race. as per the initial analysis) to the overall clinical cohort, by All markers and covariates were to be treated as categorical treatment group and overall. An exploratory estimate of the variables in the analysis. Protein biomarkers were to be response rates in each treatment group was presented with dichotomized. The cut-points for protein markers were to be 2-sided 95% confidence intervals. The number and percent optimized based on enrichment of responders vs. non-re age of Subjects falling into each response category was sponders and reasonable population size. Specifically, for descriptively tabulated. each biomarker, number of responders and non-responders Analyses were done for the population overall and by (based on overall response) and their percentages in the evalu treatment group. Each of these analyses were stratified by the 10 able population were to be determined using every potential same covariates utilized for PFS assessment. cut-point listed in the Table 3. Summary tables with potential Further analyses were planned because of the number of cut-points, number of responders, number of non-responders, pair-wise comparisons that were significant prior to FDR and their percentages were to be generated for each biomar corrections and because these pairs selected unique individu ker. als with longer PFS and a trend for longer survival. These 15 analyses sought to identify a single biomarker classifier that TABLE 3 selected for a large PFS benefit and an OS benefit (or trend) with a high population frequency. The dataset was utilized Response by Cut-Point (Evaluable Population whereby discovery and confirmation test sets within the same population were defined. The study was conducted as fol Protein Marker Cutoff lows: 20S (PSMA5)% nuclear staining 20%, 50%, 70%, 90% 20S (PSMA5)% positive cytoplasmic signal 20%, 50%, 70%, 90% Subjects with no missing biomarker values (n=354) were 20S intensity cytoplasmic signal 0,1,2 assigned in a ratio of 7:3 into a discovery and confirmation set CD68 overall positive 25, 50, 75,90 using simple randomization. The balance demographic fac CD68 positive follicular 25, 50, 75,90 tors and clinical covariates listed previously were confirmed 25 CD68 positive perifollicular 25, 50, 75,90 using either the t-test or Mann-Whitney test. The discovery P27% nuclei positive 20%, 50%, 70%, 90% P27 signal intensity 0,1,2 set (67%) was used for identification of biomarkers with P65 (NF-kB) % nuclear staining 0, 1, 5, 10, 20 significant association with PFS: subjects included had no P65 (NF-kB) % positive cytoplasmic signal 20%, 50%, 70%, 90% missing biomarker data. The confirmation set (33%) was used P65 (NF-kB) intensity cytoplasmic signal 0,1,2 for independent validation. Subjects with missing data were 30 included in the confirmation dataset provided that significant biomarker data identified in the discovery phase was avail All genotypes were to be considered separately in the able. Additionally, evaluable sample from China were also analysis (eg. C/C, C/TT/T): included in the confirmation set. FCGR2A (H166R, Q62R, Q62X) As in the initial analyses, if all Subjects have the same 35 FCGR3A (V212F) protein expression level for a particular biomarker, that biom PSMB1 (P11A) arker was removed from the analysis. If all subjects had the same mutation, no mutation, or 1 mutation level represented PSMB5 (R24C) 90% of the samples, then either that mutation or that gene was PSMB8 (G8R) removed from the analysis. 40 PSMB9 (R60H, V32I). In the discovery phase of biomarker combination analysis, Evaluation and Ranking of Single-Marker Associations with all Subjects that had missing values for any biomarker speci PFS (Discovery set) fied in Section 2 were excluded. Samples with missing data The initial step of the analysis was selection and ranking of were included in the confirmation set provided that the miss markers that were to be used in Subsequent multiple compari ing data was not part of the associated dataset. Samples with 45 son analysis. All protein, SNPs with greater than 10% geno missing biomarker data were considered “unevaluable' and type variability, and clinical covariates listed were to be were excluded from the confirmation set, all other samples included as categorical variables. The evaluation was to not used in the discovery set were included in the confirma include the following steps: tion set. Biomarkers (including clinical covariates) that showed Biomarker outliers were not removed from the analysis. 50 improvement of PFS from prior single-marker association Demographic and baseline covariates were to be compared analysis (p<0.2) were reported. Only analyses that were done using the t-test or Mann-Whitney test to ensure that there were no statistically significant differences between the dis using IRC review were used. covery and confirmation sets. Demographic and baseline Each biomarker and clinical covariate was evaluated by Cox characteristics were to be summarized for the discovery and 55 regression to assess the importance of biomarkers with confirmation sets and overall. Subjects excluded in the final respect to PFS. The P-values and weights/odds ratios of all analysis were not to be evaluated in this data comparison. markers in the Cox model were reported. Descriptive statistics (mean, Standard deviation, median, and To evaluate correlation among biomarkers, a pairwise corre range) were to be calculated for the baseline demographic lation matrix was generated using Spearman correlation data and comparison between the demographic and test set 60 method. Biomarkers that showed high correlation (p<0.05 were to be made to ensure there were no significant differ and correlation coefficient >0.7) were highlighted. Markers ences between them. that had high correlation were re-analyzed in a Cox regression Covariates included in the analysis were: FLIPI score (low model with their interaction terms. 0 or 1 factor, intermediate 2 factors, a3 high factors); An interim Summary of this analysis was generated for prior rituximab therapy (yes, no); time since last dose of 65 selecting markers that were used in multiple comparison anti-lymphoma therapy (s.1 year, >1 year), age, Ann Arbor analysis. The markers that were selected were based on the stage (I, II, III, IV), Number of prior lines of therapy (1, 2 and following criteria: US 9,322,066 B2 13 14 Relatively lower P-values in Cox regression for marker effect selected patient Subpopulations for other clinical endpoints On PFS was also evaluated. Performance of PFS and OS in the con PFS benefit in Vc-R arm compared to R arm that were based firmation set was evaluated by testing the association of on interaction with treatment in Cox regression or single biomarkers identified in the discovery set using the decision rules defined. The PFS of both biomarker positive and nega marker logrank tests. tive Subgroups was reported for both study arms in the con If multiple markers are highly correlated and have high ranks firmation set. P-value cutoff was not applicable due to a small from Cox regression, representative marker(s) that have the sample size, but was still reported. highest rank from Cox regression with interaction terms were Samples included in the independent confirmation set may used. Subpopulations of samples showing a large PFS benefit have missing values for biomarkers not found to associate in Vc-R arm compared to the Rarm were to be identified by 10 with PFS, however, subjects who could not be classified due an exhaustive search of “AND” combination of biomarkers. to missing values of the selected biomarkers were regarded as Specifically, subpopulations of patients were formed from "unevaluable' and were excluded. “AND” combinations of any two or three biomarkers selected Initial analyses focused on single-marker associations in section 4.6.1. The difference of PFS for the patient subsets stratified by covariates. Significant associations were found defined by the markers was evaluated using the log-rank test 15 in the single marker association analysis including CD68, with PFS as response variable and 5-fold cross-validation as PSMB1 (P11A), P65 and PSMB5 (R24C). Subsequent pair described below. wise analysis identified a biomarker pair with significantly The discovery set was randomly split into 5 subsets with longer PFS and a trend for an OS benefit. Because there were 20% of subjects in each subset. An 80% subset was formed by no data sets available for independent confirmation of this combining 4 of the 5 subsets. Using the 80% subset, subpopu finding, dataset was split into the Discovery and Confirmation lations of patients were formed from “AND” combinations of sets described above. As part of that analysis, multiple biom biomarkers. If the number of samples (N) in either Vc-R or R arker combinations were compared and other significant is <5 for the Subpopulation, the Subpopulation was skipped. combinations were identified. The association deemed to be For subpopulations with Na5 in both arms, the difference in most clinically appropriate from all analyses was the PSMB PFS for the patient subsets defined by the two markers was 25 P11A heterozygote in combination with CD68 Low (0-50%) evaluated using the log-rank test. A looser P-value cutoff was expression. Sub-populations of clinical interest are Subjects applied because of the exploratory nature of this analysis. The with high tumor burden, prior rituximab and one or two prior PFS benefit was subsequently tested on the remaining 20% lines of therapy. subset. Pair-wise combinations of markers were conducted using The remaining 4 cross validation sets were tested similarly 30 the stratified log rank test for each potential pair to determine (P value cutoff was not applied due to small sample size). differences in PFS between the Vc-R and rituximab only Specifically, a different 20% subset from above and a new treatment groups. One-hundred and two biomarker pairs had 80% subset formed with remaining subjects was used to a log rank p-0.05. Of these, 97 pairs had a >1% population repeat logrank tests until all 5 20% subsets were tested. The frequency. Fourteen pairs also showed a PFS improvement of proportion of iterations with large PFS benefit in both 90% 35 >6 months. In this analysis, 1,140 pair-wise comparisons subset and 10% subset were reported. were made (covariates were paired with each individual Biomarker combinations were merged and evaluated again marker to supplement the analyses). Following FDR correc for PFS benefit and statistical significance with cross valida tion, 1 pair was significant (FDR=0.051). This biomarker pair tion. For Subpopulations formed from merging of marker identified 33% of the biomarker evaluable population that had combinations, if the size of the subpopulation is s10% of the 40 a 7.5 month PFS advantage when treated with Vc-R com discovery set, no statistical test will be performed. Exhaustive pared with rituximab alone (Table 4) and a trend for better OS merging of marker pairs and assessment of PFS benefit was (p=0.055, HR: 0.426 (0.174, 1.046. This pair is composed of performed. Only marker combinations or merging of marker PSMB1 P11A (C/Gheterozygote) and CD68 Low expression combinations that were significant were reported to save defined as 0-50 positively stained cells. Following, this pair computational time. Results were saved after each iteration. 45 will be referred to as the biomarker positive subgroup. The For top ranked marker combinations, decision rules for biomarker negative subgroup does not have this biomarker defining selected patient Subpopulations were established pair and has a different PSMB1 genotype and CD68 expres using Classification and Regression Tree. Association of the sion level. TABLE 4

Comparison of Biomarker Positive Population With Biomarker Negative Population for PFS and OS

Biomarker Positive Biomarker Negative Total (N = 118) (N = 238) (N = 356)

Vc - R. Rituximab Vc - R. Rituximab Vc - R. Rituximab

PFS N 57 61 118 120 175 181 Median (months) 16.6 9.1 12.5 12.5 13.6 11.3 95% CI (0.26-0.639) (0.759-1425) (0.621-1.032) p-value O.OOO1 O.8097 O.O855 HR O.407 1.04 O.801 US 9,322,066 B2 15 16 TABLE 4-continued Comparison of Biomarker Positive Population With Biomarker Negative Population for PFS and OS Biomarker Positive Biomarker Negative Total N = 118 N = 238 N = 356 Vc - R. Rituximab Vc - R. Rituximab Vc - R. Rituximab

OS N 57 61 118 120 175 181 Median (months) NA NA NA NA NA NA 95% CI (0.174-1.046) (0.617-1.658) (0.527-1.239) p-value 0.0550 O.964.5 O.3270 HR O.426 1.011 O.808 Biomarker positive = PSMBP11A heterozygote and CD68 “Low' biomarker pair, Biomarker negative = all subjects without this pair, Vc - R = Bortezomib +Rituximab, PFS = progression free survival, OS = overall survival, CI= confidence interval, HR = hazard ratio

Importantly, biomarker positive subgroup (PSMB1 P1 1A the significance is p=0.2525 and HR=0.673 while subjects heterozygote and CD68 Low expression) also had a signifi that are CD68 positive (0-50), the significance is p=0.0714 cantly better overall response rate 73.7% for those treated and HR=0.615. with Vc-R compared to 47.5% with Ralone (p=0.0077), and Subjects treated with Velcade+rituximab had a trend for a longer time to next treatment (p=0.0013) and duration of 25 better OS when positive for CD68 low (0-50) and stratified by treatment free interval (p=0.0017). any FLIPI score, by tumor burden, by Ann Arbor score, by Similar AE profiles were observed in the biomarker posi age, by region, by sex, and by race. Subjects treated with tive and biomarker negative populations. Similar treatment Velcade+rituximab had a trend for better OS when positive exposure was observed in the biomarker positive and biom for PSMB1 P1 1A heterozygote and stratified by any FLIPI arker negative populations. Subjects treated with rituximab in 30 score, by tumor burden, by Ann Arbor score, by age, by the biomarker positive population had a median dose of 2941 region, by sex, and by race. mg/m compared to 2940 mg/m in the biomarker negative Following cross validation, the pair previously described population. Subjects treated with Vc-R in the biomarker posi (CD68 Low and PSMB P1 1A) was found to be significant in tive population had a median dose of 31.1 mg/m compared to the smaller discovery cohort (p=0.0003, 14.2 months for 30 mg/m in the biomarker negative population. Total number 35 Vc-R vs 8.5 months for R, HR=0.4 (0.24, 0.67)). There was of doses, duration of exposure, dose intensity, relative dose still a trend for longerOS in the biomarker positive population intensity and maximum number of cycles received also (p=0.1291), HR=0.47(0.17, 1.27). Although the number of showed very similar differences. Subjects in the confirmation cohorts is Small, the positive Subjects treated with Bortezomib-i-rituximab maintained trend in both PFS and OS was found to be maintained. In longer PFS when biomarker positive and stratified by any 40 confirmation cohort 1 (n=106), subjects treated with Vc-R FLIPI score, by tumor burden, by Ann Arbor score, by age, by had approximately 18.2 mo PFS while those treated with R region, by sex, and by race. In patients with higher risk and alone had 9.5 months PFS (p=0.0817, HR=0.44). In confir poor prognosis, e.g. high tumor burden, medium or high mation cohort 2 (n=426), subjects treated with Vc-R had 13.9 FLIPI, older than 65, or with prior rituximab treatment, months median PFS while those treated with R had 9.5 greater PFS improvement were observed in patients when 45 months median PFS (p=0.0878, HR=0.49). The trend in OS biomarker positive comparing to when biomarker negative. was also maintained Longer PFS was maintained regardless of time from last It was of interest to determine the individual contributions treatment or number of previous treatments by prior ritux of each biomarker to the PFS benefit found in subjects with imab or by number of rituximab treatments less than or equal both biomarkers (PSMB1 P11A heterozygote and CD68 to 2. 50 Low). Within the biomarker positive population, subjects Subjects treated with Bortezomib-i-rituximab maintained with PSMB1 P11A (C/G) had 16.6 months median PFS when longer overall survival when biomarker positive and stratified treated with Vc-R compared to 9.5 months median PFS when by FLIPI score, by tumor burden, by Ann Arbor score, by age, treated with R alone, showing a 7.1 month PFS advantage by region, by sex, and by race. Subjects treated with Velcade-- when the combination was used. This benefit was not seen in rituximab maintained longer PFS when positive for CD68 55 biomarker negative subjects with Vc-R. Consistent with prior low (0-50) and stratified by any FLIPI score, by tumor burden, publication on CD68 and rituxan, this study shows that sub by Ann Arbor score, by age, by region, by sex, and by race. jects with higher CD68 expression do better on rituximab When CD68 high, they did better on rituximab alone as alone (16.2 months median PFS) than those with lower CD68 expected. Subjects treated with Velcade-rituximab main expression (9.3 months median PFS). However, these sub tained longer PFS when positive for PSMB P1 1A heterozy 60 jects with low CD68 (0-50) had significantly longer PFS (14 gote and stratified by any FLIPI score, by tumor burden, by months median) when treated with the combination com AnnArbor score, by age, by region, by sex, and by race. When pared to similar patients treated with R alone (9.3 months CD68 high, they did better on rituximab alone as expected. median); this is a 4.7 month PFS advantage (HR-0.64 (95% A trend for a longer OS was found for biomarker positive CI: 0.475-0.864). subjects (p=0.055, HR 0.426. When the biomarker contribu 65 These results suggest that patient subgroups can be identi tions are examined individually, this trend is less distinguish fied prior to therapy that have significantly longer PFS and a able. For subjects that are PSMB1 P11 AC/G heterogygotes, trend for better OS when treated with combination Vc-R US 9,322,066 B2 17 18 compared to R alone. One biomarker pair that maintains found to be synergistic and present in a high proportion of the significance after multiple comparison corrections study cohort. PSMB1 P1 1A is a polymorphism found in the (FDR=0.051) includes a proteasomal subunit SNP PSMB1 leader sequence of the proteasome PSMB1 gene (Chen P1 1A heterozygote and CD68 with low (0-50) expression. 1996). The leader sequence is responsible for appropriate 5 subunit assembly and it has been reported that alterations of Three hundred fifty six subjects were evaluable for both of charged amino acids reduces the efficiency of subunit assem these biomarkers within this study. Additionally, this biom bly (Schmidt 1999). Once assembled autocatalysis allows arker pair had a high population frequency with approxi removes the leader sequence. The second biomarker in the mately one third of the evaluable subjects having both of these pairis CD68 expressing tumor infiltrating macrophages. Pre biomarkers. Subjects with this biomarker pair had a longer vious reports have shown that high levels of CD68 associate PFS interval when treated with Vc-R (16.6 months) compared 10 with better response to rituximab (Taskinen 2007). to rituximab alone (9.1 months) demonstrating a PFS benefit Importantly, this study has shown that subjects with low of approximately 7.5 months with the combination. This levels of CD68 expression have longer PFS and better overall group also appears to have an OS benefit with Vc-R that response to Vc-R compared to R alone especially when clearly trends toward significance (HR: 0.426 (0.174-1.046) present with PSMB1 P1 1A heterozygote as discussed above. P=0.0550), a higher ORR (73.7% on Vc-R vs 47.5% on 15 Following cross validation, the pair previously described rituximab alone, p=0.0077), a longer time to next treatment (CD68 Low and PSMB P1 1A) was found to be significant in interval (33.1 mo on Vc-R vs 14.8 moon rituximab alone, the smaller discovery cohort (p=0.0003, 14.2 months for p=0.0013) and a longer duration of treatment free interval Vc-R vs 8.5 months for rituximab alone, HR=0.4 (0.24. (27.8 mo vs 10.1 mo, p=0.0017). 0.67)). There was still a trend for longer OS in the biomarker Importantly, the CD68/PSMB P1 1A biomarker positive positive population (p=0.1291), HR=0.47(0.17, 1.27). cohort was representative of the overall trial population with Although the number of subjects in the confirmation cohorts regard to demographics and clinical characteristics. Of note, was small, the positive trend in both PFS and OS was found to approximately half of this cohort was represented by subjects be maintained. In confirmation cohort 1 (which excludes with high risk disease and poor prognostic features. In subjects from China; n=106), subjects treated with Vc-R had patients with higher tumor burden (-54%), medium or high 25 approximately 18.2 mo PFS while those treated with Ralone FLIPI (-76%), older age (>65) (-25%), with prior Rituximab had 9.5 months PFS (p=0.0817, HR=0.44). In confirmation treatment (~46%), or with two (-26%) or more than two cohort 2 (which includes China subjects and subjects with (-30%) lines of prior therapy the PFS benefit of the CD68/ missing data; n=126), subjects treated with Vc-R had 13.9 PSMB P1 1A biomarker positive cohort appears to be main months median PFS while those treated with R had 9.5 tained. These patients are generally regarded as "high risk” 30 months median PFS (p=0.0878, HR=0.49). The trend in OS with limited treatment options and these preliminary findings was also maintained. Suggest that biomarker positive Subjects with high risk fea tures may benefit from Vc-R. EXAMPLE 2 Polymorphisms in PSMB1 and PSMB5 were found to have significant associations with clinical endpoints as single 35 Single predictors which correlated with a positive response markers. The combination of PSMB1 (P11A with CD68 was are shown in Table 5. TABLE 5

Single predictors.

PFS Marker Vc + R vs. R N Logrank % N in Marker A Subtype median days Vc + RWS.R P-value ITT

CD68 OVERALL O-2S 1.5 mo vs 10.6 mo 40 vs. 44 O422 12.4 POSITIVE 0.9 mo PFS improvement CD68 OVERALL 26-SO 2.1 mo vs. 9.3 mo 114 wS 108 O.O588 32.9 POSITIVE 2.8 mo PFS improvement CD68 POSITIVE O-2S 4.1 mo vs. 9.3 mo SOws 60 O.O934 16.3 FOLLICULAR 4.8 mo PFS improvement CD68 POSITIVE 26-SO 3.4 mows 9.1 mo 84 wS 91 O.O289 25.9 FOLLICULAR 4.3mo PFS improvement P65 INTENSITY <=1- 1.6 mo vs. 9.3 mo 41 vs 43 O.2455 12.4 CYTOPLASMIC 2.3 mo PFS improvement SIGNAL PSMB1 P11A CG 4 mo vs. 9.3 mo 115 vs 127 0.0218 35.9 4.7 mo improvement PSMBSR24C CT 7.6 mo vs. 9.3 mo 41 vs 41 O4O16 12.1 8.3 mo improvement US 9,322,066 B2 19 20 EXAMPLE 3 Chen S (2010), Blank J. L. Peters T. Liu XJ, Rappoli D M, Pickard M. D. Menon S. Driscoll D. L. Lingaraj T. Predictor pairs which correlated with a positive response Burkhardt A L, Chen W. Garcia K, Sappal D S. Gray J, are shown in Tables 6 and 7. Hales P. Leroy P.J. Ringeling J, Rabino C. Spelman JJ, TABLE 6 Significant marker pairs.

PFS Vc - Rws. R N Logrank Marker A Marker B median month Vc - R vs. R P-value FDR

PSMB5.R24C P65 INTENSITY 27 mo vs. 10.4mo S vs. 7 O.0439 O.489 C.T CYTOPLASMIC SIG- 6.6 mo improvement NAL <=1+ PSMB1 P11 A 20S 90 POSITIVE 8.9mo vs. 9.5 mo SO vs SO O.O145 0.447 C.G. CYTOPLASMIC SIG- 9.4 improvement NAL: >90 PSMB1 P11 A CD68 POSITIVE 6.6 mo vs. 9.1 mo 57 vs 61 O.OOO1 O.OS1 C.G. FOLLICULAR: O-SO 7.5 mo improvement PSMB1 P11 A CD68 POSITIVE 6.6 mo vs. 9.2 mo 24 vs 28 O.O365 0.471 C.G. PERIFOLLICULAR: >50 7.4 improvement PSMB9R6OH P65% NUCLEAR 6.2 mo vs. 9.5 mo 35 vs 28 O.O3O3 O4S5 GiG STAINING: -O 6.7 improvement PSMBSR24C CD68 POSITIVE 3.7 mo vs. 7.2 mo 18ws 21 O.O220 O.447 C.T FOLLICULAR: O-SO 6.5 mo improvemen HITumor BD CD68 OVERALL 22.8 mo vs. 16 mo 64 wS 68 O.O177 O.447 NO POSITIVE: O-50 6.8 mo improvemen HITnor BD CD68 POSITIVE 20.5 mo vs. 13.8 mo 64 wS 66 O.O310 O4S5 NO FOLLICULAR: O-SO 6.7 mo improvemen Prior RX: 1 CD68 POSITIVE 18.2 mo vs. 9.3 mo 63 wS 69 O.O129 O.447 FOLLICULAR: O-SO 8.9 mo improvemen PSMB1 P11 A Time since last 18.2 mo vs. 10.7mo 72 vs 74 O.O.198. O.447 C.G. Rx: >1 year 7.5 mo improvemen Prior Rittlx CD68 POSITIVE 15.9 mo vs. 9.2 mo 73 vs 86 O.OO66 0.437 NO FOLLICULAR: O-SO 6.7 mo improvemen PSMB1/P11A Age group: <=65 15.3 mo vs. 9.2 mo 86 wS 96 O.OO71 O.437 CFG 6.1 mo improvemen Sex 2OS 9 NUCLEAR 13.7 mo vs. 7.7 mo 63 wS 48 O.OOSO 0.437 MALE STAINING: -2O 6 mo improvement Race Group 2OS 9 NUCLEAR 11.4 mo vs. 3.8 mo 11 vs 7 O.O320 O4S5 OTHER STAINING: -2O 7.6 mo improvemen PSMB1 P11 A PSMBSR24C 13.7 mo vs. 7.8 mo 7 vs. 7 O.O221 O.4468 C.G. CT 5.9 mo improvemen

TABLE 7 40 Morganstem J P Lightcap E S. Genome-wide siRNA screen for modulators of cell death induced by proteasome Significant marker pairs inhibitor bortezomib. Cancer Res 2010; 70(11): 4318 PFS 4326. Vc - R vs. R Logrank Clinical Study Protocol 26866138-LYM3001. A randomized, Combination median month P-value 45 open-label, multicenter study of VELCADE with ritux P65 Cytoplasmic signal -90% & 23.6 vs. 10.6 mo (13 mo) 0.0132 imab or rituximab alone in subjects with relapsed or refrac 1 prior treatment 16 vs. 8.9 mo (7.1 mo) l.S. tory, rituximab naive or sensitive follicular B-cell non CD68 Pos Follic (O-50) & 14.2 vs 8.5 mo (5.7 mo) 0.0025 Hodgkin’s lymphoma. Document No. EDMS-PSDB P11AC/G|** 14.4 vs. 9.2 mo (5.2 mo) n.s. 4649082:4.0; Johnson & Johnson Pharmaceutical 50 Research & Development (19 May 2006). Arif A (2009), Jamal S, Mushtaq S. Ahmed S. Mubarik A. Clinical Study Report 26866138-LYM3001. A randomized, Frequency of bcl-2 gene rearrangement in B-Cell Non open-label, multicenter study of VELCADE with ritux Hodgkin’s lymphoma. Asian Pacific J Cancer Prev 2009: imab or rituximab alone in subjects with relapsed or refrac 10(2): 237-240. 55 tory, rituximab naive or sensitive follicular B-cell non Binstadt BA (2003), Geha RS, Bonilla F.A. IgGFc receptor Hodgkin’s lymphoma. Document No. EDMS-ERI polymorphisms in human disease: Implications for intra 16225335:1.0; Johnson & Johnson Pharmaceutical venous immunoglobulin therapy. J Allergy Clin Immunol Research & Development (9 Feb. 2011). 2003; 111(4): 697-703. Dave SS (2004), Wright G, Tan B. Rosenwald A. Gascoyne Cartron G (2002), Dacheux L., Salles G, Solal-Celigny P 60 R D, Chan W C, Fisher R I, Braziel RM, Rimsza L. M., Bardos P. Colombat P. Watier H. Therapeutic activity of Grogan TM, Miller T P Leblanc M, Greiner TC, Weisen humanized anti-CD20 monoclonal antibody and polymor burger D D, Lynch J C, Vose J. Armitage JO, Smeland E B, phism in IgG Fc receptor FcgammaRIIIa gene. Blood Kvaloy S. Holte H, Delabie J. Connors J. M. Lansdorp PM, 2002; 99(3): 754-758. Ouyang Q, Lister T. A. Davies A. J. Norton A. J. Muller Chen P (1996), Hochstrasser M. Autocatalytic subunit pro- 65 Hermelink HK, Ott G, Campo E. Montserrat E. Wilson W cessing couples active site formation in the 20S protea H. Jaffe E S, Simon R, Yang L, Powell J, Zhao H, Gold some to completion of assembly. Cell 1996: 86: 961-972. schmidt N. ChioraZZiM, Staudt L. M. Prediction of survival US 9,322,066 B2 21 22 in follicular lymphoma based on molecular features of growth arrest and apoptosis in human glioblastoma multi tumor-infiltrating immune cells. NEJM 2004; 351 (21): forme (GBM). Oncogene 2005; 24(3): 344-354. 2159-21.69. Fischer U (2005), Schulze-Osthoff K. New approaches and EXAMPLE 4 therapeutics targetingapoptosis in disease. Pharmacol Rev 5 2005; 57(2): 187-215. Appendix 2 presents an overall Summary of the single Goy A (2010), Bernstein S, McDonald A. Pickard M, Shi H, marker associations with clinical endpoint other than PFS and Fleming M. Bryant B, Trepicchio W. Fisher R. Boral A, stratified by clinical covariates from the previous examples. Mulligan G. Potential biomarkers of bortezomib activity in Appendix 3 outlines the data for all significant pair-wise mantle cell lymphoma from the phase 2 PINNACLE trial. 10 combinations from the previous examples. Note: Leukemia & Lymphoma 2010; 51(7): 1269-1277. Selected=Biomarker positive, Not Selected=Biomarker Karin M(2002), Cao Y, Florian GR, Li ZW. NF-kB in cancer: negative. From innocent bystander to major culprit. Nature Rev Can EXAMPLE 5 cer 2002; 204): 301-310. 15 Keats J J (2007), Fonseca R, Chesi M. Schop R. Baker A, Chng Wi, Van Wier S, Tiedemann R, Shi CX, Sebag M, The VISTA study was an open-label, randomized study of Braggio E. Henry T. Zhu Y X, Fogle H, Price-Troska T. VELCADE/Melphalan/Prednisone versus Melphalan/Pred Altmann G. Mancini C, Brents LA, Kumar S. Greipp P. nisone in Subjects with previously untreated multiple Dispenzieri A, Bryant B. Mulligan G. Bruhn L. Barrett M. myeloma. San Miguel, N. Engl. J. Med. 2008 Aug. 28; 359 Valdez R. Trent J. Stewart A. K. Carpten J. Bergsagel PL. (9):906-17. The primary efficacy objective of this study was Promiscuous mutations activate the noncanonical NF-kap to determine whether the addition of VELCADE (Bort paB pathway in multiple myeloma. Cancer Cell 2007: eZomib for Injection) to standard melphalan/prednisone (MP) 12(2): 131-144. therapy improves the time to disease progression (TTP) in Mitsiades N (2002), Mitsiades CS, Poulaki V. Chauhan D, 25 Subjects with previously untreated multiple myeloma. Fanourakis G, Gu X, Bailey C. Joseph M. Libermann T.A. The exploratory objectives in biomarker analysis from the Treon S P Munshi NC, Richardson P G, Hideshima T, VISTA study were to identify patient populations that are Anderson KC. Molecular sequelae of proteasome inhibi more or less likely to respond to VELCADE/Melphalan/ tion in human multiple myeloma cells. Proc Natl Acad Sci Prednisone versus Melphalan/Prednisone alone by: USA 2002; 99(22): 14374-14379. 30 Confirming the finding (from lymphoma studies) of a Mulligan G (2007), Mitsiades C, Bryant B, Zhan F. Chng W single marker association of PSMB1 P11A and PSMB5 J. Roels S. Koenig E. Fergus A, Huang Y. Richardson P. R24C with PFS and OS. Trepicchio W L. Broyl A. Sonnveld P. Shaghnessy J D. Association with other clinical endpoints including: time Bergsagel P. L. Schenkein D. Esseltine D L, Boral A, to progression (TTP), complete response (CR), Overall Anderson KC. Gene expression profiling and correlation 35 response rate, time to response and duration of response. with clinical outcome in clinical trials of the proteasome Association of PSMB1 P11A and PSMB5 R24C individu inhibitor bortezomib. Blood 2007: 109(8): 3177-3188. ally or in combination with TTP, PFS, and OS were estimated Schmidt M(1999), ZantopfD, Kraft R, Kostka S. Preissner R, using the log rank test between treatment groups for biomar Kloetzel P. M. Sequence information within proteasomal ker positive and negative populations. The overall biomarker prosequences mediates efficient integration of B-subunits 40 population by treatment arm had similar associations made. into the 20S proteasome complex.J Mol Biol 1999; 288(1): Medians of TTP, PFS, and OS, difference in median TTP, 117-128. PFS, and OS between treatment groups, log rank P-value, Shin HM (2006), Minter LM, Cho OH, Gottipati S, Fauq A hazard ratio and its 95% confidence intervals and frequencies H, Golde T E, Sonenshein G. E. Osborne B. A. Notchl of events were reported. Kaplan-Meier plots were presented augments NF-kappaB activity by facilitating its nuclear 45 for each biomarker. When positive associations were found retention. EMBO 2006: 25(1): 129-138. for TTP, OS or PFS, then the other clinical endpoints were Taskinen M (2007), Karjalainen-Lindsberg ML, Nyman H, tested with similar methods and output. For ORR. Fishers Eerola L. M. Leppa S. A high tumor-associated macroph exact test was used. The number and percentage of Subjects age content predicts favorable outcome in follicular lym falling into each response category were descriptively tabu phoma patients treated with rituximab and cyclophospha 50 lated. mide-doxorubicin-Vincristine-prednisone. Clin Cancer Table 8 shows the progression free survival benefit of 5.3 Res 2007: 13(19): 5784-5789. months with VELCADE-MP vs. MP alone in patients with Wojcik C (2002). Regulation of apoptosis by the ubiquitin PSMB1 P11A (C/G) marker. and proteasome pathway. J Cell Mol Med 2002; 6(1): 25-48. 55 Wu J (1997), Edberg J C, Redecha PB, Bansal V. Guyre PM, Biomarker Biomarker Coleman K. Salmon J. E. Kimberly RP. A novel polymor Positive Negative Total phism of FcgammaRIIIa (CD16) alters receptor function Statistic Vmp mp Vmp mp Vmp mp and predisposes to autoimmune disease. JClin Invest 1997: Event 39.82 61.89 47/87 68.94 86,169 129, 183 100(5): 1059-1070. 60 Total (47.6) (68.5) (54.0) (72.3) (50.9) (70.5) Yeung BH (2006), Huang DC, Sinicrope FA. PS-341 (bort (%) eZomib) induces lysosomal cathepsin b release and a Median 20.3 mo 15 mo 17.7 mo 11.2 mo 19 mo 12.6 mo caspase-2 dependent mitochondrial permeabilization and (95% CI) (556, (253, (459, (254, (533, (279, 873) 511) 610) 436) 662) 463) apoptosis in human pancreatic cancer cells. J Biol Chem HR 0.44 O.62 O.S4 2006: 281 (17): 11923-11932. 65 (Vmp vs (0.29, (0.43, (0.41, Yin D (2005), Zhou H. Kumagai T, Liu G. Ong JM, Black K mp) 0.67) 0.90) 0.71) L. Koeffler H. P. Proteasome inhibitor PS-341 causes cell US 9,322,066 B2

-continued Biomarker Positive Biomarker Negative Total B k B k lomar e lomar e Statistic Vmp mp Vmp mp Vmp mp Positive Negative Total 5 Event Total 47.82 65.89 42.87 SS,94 89,169 120,183 Statistic Vmp mp Vmp mp Vmp mp (%) (57.3) (73.0) (48.3) (58.5) (52.7) (65.6) Median 52 39 58 46 56 42 HR p- <.0001 O.O117 <.OOO1 (95% CI) (1322, (893, (1318, (965, (1372, (1014, Wall 1845) 1412) —) 1745) 1845) 1470) le HR O.63 O.81 0.72 (Vmp vs 10 Vmp vs (0.44, (0.54, (0.55, mp) mp) 0.92) 1.22) 0.95) HR p- O.O167 O.3153 O.O179 Value (Vmp vs Table 9 shows the overall survival benefit of 6 months with mp) VELCADE-MP vs. MP alone in patients with PSMB1 P1 1A (C/G) marker. APPENDIX 2, TABLE 2.1 Overall Survival (OS) by Protein Expression and by Covariate, IRC Review (Significant Ips 0.05), Frequency of a 10% or Higher Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal 2OS 90 POSITIVE 2 Prior 8.28 1,10 3f4 809 122 CYTOPLASMIC Lines o (0.85, 80.83) SIGNAL: O-2O Therapy 2OS 90 POSITIVE 2 Prior O.42 14,3S 1205 9.f40 122 CYTOPLASMIC Lines o (0.17, 1.00) SIGNAL: 95-100 Therapy CD68 OVERALL 2 Prior O.33 13.26 1205 7.30 111 POSITIVE: 26-50 Lines o (0.12, 0.87) Therapy CD68 POSITIVE 2 Prior O.30 7.16 1205 5/27 98 PERIFOLLICULAR: Lines O (0.09, 1.04) 26-SO Therapy P659. POSITIVE 2 Prior O.42 1952 9.f44 125 CYTOPLASMIC Lines o (0.19, 0.97) SIGNAL: -90% Therapy P65 INTENSITY 2 Prior O.35 1955 7.43 125 CYTOPLASMIC Lines o (0.14, 0.88) SIGNAL:2- Therapy CD68 OVERALL No High O.21 14.54 3f46 204 POSITIVE: 26-50 Tumor (0.06, 0.72) Burden CD68 POSITIVE No High O.11 10:41 1.33 182 PERIFOLLICULAR: Tumor (0.01, 0.82) 26-SO Burden P279% NUCLEI No High 3.91 3,29 6.17 215 POSITIVE: O-20 Tumor (0.98, 15.69) Burden 2OS INTENSITY intermediate 5.43 2.f48 9.f41 168 CYTOPLASMIC FLIPI (1.17, 25.14) SIGNAL: s2- Score P65 INTENSITY intermediate 6.84 1.17 5,13 170 CYTOPLASMIC FLIPI (0.80, 58.74) SIGNAL: s1- Score CD68 POSITIVE No Prior O.21 11.36 3.25 1343 210 FOLLICULAR: Rituximab (0.05, 0.97) O-2S Therapy CD68 POSITIVE sé5 years O.29 14.f46 S.41 292 FOLLICULAR: old (0.10, 0.83) O-2S

2OS INTENSITY Male O.30 10,27 8,62 204 CYTOPLASMIC (0.12, 0.77) SIGNAL: e3+ US 9,322,066 B2 25 26 APPENDIX 2, TABLE 2.1-continued Overall survival (OS) by Protein Expression and by Covariate, IRC Review (Significant Ips 0.05), Frequency of a 10% or Higher

Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal

CD68 POSITIVE Male O.29 11.30 539 169 PERIFOLLICULAR: (0.10, 0.83) 26-SO

P279% NUCLEI Male O.09 412 114 2O2 POSITIVE: 60-70 (0.01, 0.93)

CD68 OVERALL Ann Arbor 6.24 2.21 6.13 1078 144 POSITIVE: 51-75 Stage III (1.25, 31.02) 2OS 90 POSITIVE Ann Arbor 7.22 1.15 411 1103 236 CYTOPLASMIC Stage IV (0.80, 65.02) SIGNAL: O-2O 2OS 90 POSITIVE Ann Arbor OSO s 25.56 1205 1975 1343 236 CYTOPLASMIC Stage IV (0.27, 0.91) SIGNAL: 95-100

O.O1 O.1 1 10 100

APPENDIX 2, TABLE 2.2 OS by Germline Genetic Variant nd by Covariate, IRC Review (Significant Ips 0.05), Frequency of e10% or Higher HR R R Vc - R Vc - R Marker: Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal PSMB9,R6OH: 3 Prior 6.92 1.6 5/7 846 85 AA Lines of (0.79, 60.79) Therapy PSMBSR24C: Intermediate O.10 4f10 971 1.16 186 C.T FLIPI (0.01, 0.97) Score PSMBSR24C: Ann Arbor O.22 9/15 717 5,28 270 C.T Stage IV (0.07, 0.66)

O.O1 O.1 1 10 100

APPENDIX 2, TABLE 2.3 OS by Somatic Mutation and by Covariate, IRC Review (Significant Ips 0.05), Frequency of e10% or Higher Marker: HR R R Vc - R Vc -R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal NOTCHSX28DEL: 2 Prior 0.44 20, S6 10.48 109 MIND Lines of (0.20, 0.97) Therapy NOTCH/P2513L: No High O.14 7.26 1.21 163 MD Tumor (0.02, 1.15) Burden BCL2/P59L: MD High O.17 5.6 282 4f14 172 Tumor (0.04, 0.65) Burden BCL2/P59L: MD High O.21 66 282 4f7 843 125 FLIPI (0.05, 0.86) Score US 9,322,066 B2 27 28 APPENDIX 2, TABLE 2.3-continued OS by Somatic Mutation and by Covariate, IRC Review (Significant Ips 0.05), Frequency of a 10% or Higher Marker: HR R R Vc - R Vc -R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal BCL2fPS9L: MD NO Prior O.12 5/5 174 4f13 178 Rituximab (0.03, 0.48) Therapy BCL2fPS9L: MD Rest of O.22 4.5 174 5, 12 152 World (0.06.0.84 NOTCHP2513Li: -65 O.30 8, 13 795 4f13 92 MD years old (0.09, 1.03) BCL2P59L: MD Female O.26 6.8 536 4f14 18O (0.07, 0.92) BCL2iR106H: Male O.25 8.10 318 4f12 169 MD (0.07, 0.84) NOTCH/Q246OX: Male O.11 3.3 595 5, 12 142 MD (0.02, 0.70) NOTCHSX28DEL: Male 0.57 2675 22.97 18O MIND (0.32, 1.00)

BCL2fPS9L: MD Ann O.12 5/5 174 4f12 149 Arbor (0.03, 0.50) Stage IV

O.O1 O.O2 0.05 0.1 O2 (0.5 1

APPENDIX 2, TABLE 2.4 Time to Progression (TTP), by Protein Expression and by Covariate, IRC Review Significant Ips 0.05l. Frequency of e10% or Higher Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal CD68 POSITIVE No O.66 s 65.91 277 48.84 414 387 FOLLICULAR: 26-50 Subgroup (0.45, 0.96) P659. POSITIVE No 0.75 s 139,204 334 116,186 414 470 CYTOPLASMIC Subgroup (0.59, 0.96) SIGNAL: -90% CD68 POSITIVE Prior O.S1 H 26,39 349 18,40 881 176 FOLLICULAR:26-50 Line of (0.27, 0.94) Therapy CD68 POSITIVE Prior O.24 13,18 275 7, 16 716 174 PERIFOLLICULAR: Line of (0.09, 0.65) >75 Therapy P659. POSITIVE Prior 0.67 s 56.89 349 48.88 SO6 2O3 CYTOPLASMIC Line of (0.45, 0.98) SIGNAL: -90% Therapy 2OS 9/o NUCLEAR 2 Prior O 41 11.13 239 13,22 431 122 STAINING: 30-50 Lines of (0.18, 0.95) Therapy CD68 OVERALL 2 Prior O.22 6.8 142 4f7 771 111 POSITIVE: O-25 Lines of (0.05, 0.90) Therapy CD68 POSITIVE 2 Prior O.13 7/8 70 4f7 771 98 PERIFOLLICULAR: Lines of (0.03, 0.64) O-2S Therapy

CD68 POSITIVE 3 Prior 9.40 1.f4 8.8 276 60 FOLLICULAR: 51-75 Lines of (1.13, 78.09) Therapy US 9,322,066 B2 29 30 APPENDIX 2, TABLE 2.4-continued Time to Progression (TTP), by Protein Expression and by Covariate, IRC Review Significant Ips 0.05l. Frequency of a 10% or Higher Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal P27 SIGNAL 4 Prior 3.02 9/15 348 8.10 144 32 NTENSITY: e2- Lines of (1.09, 8.34) Therapy CD68 OVERALL No High O.SO 37/54 357 22.f46 881 204 POSITIVE: 26-50 Tumor (0.29, 0.86) Burden CD68 POSITIVE No High O.SO 29:44 351 1739 881 182 FOLLICULAR:26-50 Tumor (0.27, 0.93) Burden CD68 OVERALL High FLIPI O.S8 H 36.45 277 35,52 366 181 POSITIVE: 26-50 Score (0.36, 0.94)

CD68 POSITIVE High FLIPI 0.57 H 31,37 205 2S,36 347 156 FOLLICULAR:26-50 Score (0.33, 0.97)

P659, POSITIVE High FLIPI O.65 ke 61.83 239 53,77 358 193 CYTOPLASMIC Score (0.45, 0.94) SIGNAL: -90% P65 INTENSITY High FLIPI 0.67 e 60.80 275 51.76 358 193 CYTOPLASMIC Score (0.46, 0.98) SIGNAL:2-

P279. NUCLEI intermediate 3.34 6.13 513 10:11 215 165 POSITIVE: 60-70 FLIPI (1.20, 9.30) Score

P659, POSITIVE intermediate 2.77 Sf 16 567 1S 19 351 170 CYTOPLASMIC FLIPI (1.00, 7.64) SIGNAL: soyo Score P65% NUCLEAR No Prior O.69 s 77,102 322 55/92 426 255 STAINING: O Rituximab (0.49, 0.97) Therapy P659, POSITIVE No Prior O.68 s 80.112 345 61.1OS 463 255 CYTOPLASMIC Rituximab (0.49, 0.95) SIGNAL: -90% Therapy

CD68 OVERALL s1 year 2.66 11.18 424 11.13 2O2 173 POSITIVE: O-25 since last (1.02, 6.96) anti-lymphoma treatinent

CD68 POSITIVE >1 year O.49 23,34 357 2O33 519 235 FOLLICULAR: 0-25 since last (0.26, 0.91) anti-lymphoma treatinent P659. POSITIVE >1 year O.69 s 77,117 357 61,112 S19 288 CYTOPLASMIC since last (0.49, 0.96) SIGNAL: -90% anti-lymphoma treatinent CD68 OVERALL Rest of O.63 H 40.51 277 34,52 367 198 POSITIVE: 26-50 World (0.40, 1.00) CD68 OVERALL sé5 years O.65 e 57.77 278 SO.84 414 329 POSITIVE: 26-50 old (0.44, 0.95) CD68 POSITIVE sé5 years 0.55 s 52.67 270 34f62 406 292 FOLLICULAR:26-50 old (0.36, 0.86) CD68 POSITIVE sé5 years O46 1723 275 12.23 SO6 289 PERIFOLLICULAR: Old (0.22, 0.98) US 9,322,066 B2 31 32 APPENDIX 2, TABLE 2.4-continued Time to Progression (TTP), by Protein Expression and by Covariate, IRC Review Significantgil ps. 0.05, FrequencyC y of 10% or Higher9.

Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal P27 SIGNAL sé5 years 0.73 ke 104,153 281 91.149 406 344 INTENSITY:2- ol (0.55, 0.97) P65% NUCLEAR sé5 years 0.72 k 97,138 287 71,118 414 349 STAINING: O ol (0.53, 0.98) P659. POSITIVE sé5 years O.66 s 108,154 278 83.142 422 349 CYTOPLASMIC ol (0.50, 0.89) SIGNAL: -90%

P65 INTENSITY sé5 years O.74 ke 105,152 287 85,139 406 349 CYTOPLASMIC ol (0.55, 0.98) SIGNAL:2-

2OS 9. POSITIVE >65 years O.14 5.6 278 4f7 738 118 CYTOPLASMIC ol (0.02, 1.21) SIGNAL: O-2O

CD68 POSITIVE Female 2.38 1426 708 13,14 324 217 FOLLICULAR: S1-75 (1.11, 5.13)

2OS 9/o NUCLEAR Male 0.37 13,16 212 1S,2O 358 204 STAINING: 60-70 (0.17, 0.84)

2OS INTENSITY Male O.S8 20,27 28O 39.62 422 204 CYTOPLASMIC (0.34, 1.00) SIGNAL: e3+

CD68 POSITIVE Male O.48 22.30 239 26,39 429 169 PERIFOLLICULAR: (0.27, 0.86) 26-SO

P27 SIGNAL Male O.70 s 52/72 28O 69,101 360 2O2 INTENSITY:2- (0.48, 1.00)

P65% NUCLEAR Male 0.67 H 56,74 28O 56.84 358 2O7 STAINING: O (0.46, 0.98)

P659. POSITIVE Male O.61 s 54f72 271 65.97 414 2O7 CYTOPLASMIC (0.42, 0.87) SIGNAL: -90%

CD68 POSITIVE Other O.14 44 128 4f S 346 22 PERIFOLLICULAR: (0.01, 1.31) ' 26-SO

CD68 POSITIVE White O.S9 s 53,77 334 39,73 463 335 FOLLICULAR: 26-50 (0.39, 0.89)

P659. POSITIVE White 0.72 H 119,179 345 99.162 426 406 CYTOPLASMIC (0.55, 0.94) SIGNAL: -90%

P65 INTENSITY Ann Arbor O.26 7 11 204 7 11 464 151 CYTOPLASMIC Stage III (0.08, 0.90) SIGNAL: s1

2OS 9. POSITIVE Ann Arbor O.66 e 43 S6 278 47/75 414 236 CYTOPLASMIC Stage IV (0.43, 1.00) SIGNAL: 95-100

CD68 OVERALL Ann Arbor O.65 H 4456 277 45.68 366 222 POSITIVE: 26-50 Stage IV (0.42, 0.99)

P279% NUCLEI Ann Arbor O.43 1S 19 278 13,23 485 237 POSITIVE: 30-50 Stage IV (0.20, 0.92)

P27 SIGNAL Ann Arbor O.70 H 64,85 283 73.113 358 237 INTENSITY:2- Stage IV (0.50, 0.99) US 9,322,066 B2 33 34 APPENDIX 2, TABLE 2.4-continued Time to Progression (TTP), by Protein Expression and by Covariate, IRC Review (Significant ps. 0.05), Frequency of a 10% or Higher)

Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal

P659. POSITIVE Ann Arbor O.65 - 67187 275 66,104 366 240 CYTOPLASMIC Stage IV (0.46, 0.91) SIGNAL: -90%

O.O1 O.1 1 10 100

APPENDIX 2, TABLE 2.5 TTP by Germline Genetic Variant and by Covariate, IRC review Significant Ips 0.05l. Frequency of e10% or Higher Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal PSMB1 P11A: No O.70 82,127 288 74f11S 426 S42 C.G. Subgroup (0.51, 0.96) PSMB9, V32I: No O.8O 177,266 345 157254. 417 S42 CFC Subgroup (0.65, 0.99) PSMBSR24C: 2 Prior O.32 8.10 28O 612 534 142 C.T Lines of (0.11, 0.95) Therapy PSMB1/A171S: High O.74 110,147 273 107,149 344 296 GiG Tumor (0.57, 0.97) Burden PSMB1 I2O8N: High O.74 110,147 273 107,149 344 296 TT Tumor (0.57, 0.97) Burden

PSMB1 P11A: High O.66 O 52/72 253 44f65 358 296 C.G. Tumor (0.44, 0.99) Burden PSMB1 P193Li: High O.74 Of 147 273 O7,149 344 296 CFC Tumor (0.57, 0.97) Burden PSMB2E49X: High O.74 Of 147 273 O7,149 344 296 GiG Tumor (0.57, 0.97) Burden PSMB2/G187V: High O.74 Of 147 273 O7,149 344 296 GiG Tumor (0.57, 0.97) Burden PSMB2FL159F: High O.74 Of 147 273 O7,149 344 296 CFC Tumor (0.57, 0.97) Burden PSMB5/L206M: High O.74 Of 147 273 O7,149 344 296 CFC Tumor (0.57, 0.97) Burden PSMBSR24C: High O.39 7.21 275 1824 352 296 C.T Tumor (0.19, 0.77) Burden PSMB6/A234D: High O.74 110,147 273 107,149 344 296 CFC Tumor (0.57, 0.97) Burden PSMB8.G8R: High 0.73 104,14O 271 105,145 344 296 GiG Tumor (0.56, 0.96) Burden

US 9,322,066 B2 41 42 APPENDIX 2, TABLE 2.5-continued TTP by Germline Genetic Variant and by Covariate, IRC review Significant Ips 0.05l. Frequency of a 10% or Higher Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) HR (log scale) Evti N Median Evti N Median NTotal

PSMB1 P11A: Ann Arbor O.64 46.59 235 48.68 351 270 C.G. Stage IV (0.42, 0.96) He PSMB1 P193L: Ann Arbor 0.72 95,127 278 93.143 360 270 CFC Stage IV (0.54, 0.95) He PSMB2E49X: Ann Arbor 0.72 95,127 278 93.143 360 270 GiG Stage IV (0.54, 0.95) He PSMB2G187V: Ann Arbor 0.72 95,127 278 93.143 360 270 GiG Stage IV (0.54, 0.95) H PSMB2L159F: Ann Arbor 0.72 95,127 278 93.143 360 270 CFC Stage IV (0.54, 0.95) He PSMBSL206M: Ann Arbor 0.72 95,127 278 93.143 360 270 CFC Stage IV (0.54, 0.95) He PSMB6A234D: Ann Arbor 0.72 95,127 278 93.143 360 270 CFC Stage IV (0.54, 0.95) He PSMB6P107A: Ann Arbor O.74 He- 92,124 279 92,140 358 270 CFC Stage IV (0.55, 0.99) PSMB8.G8R: Ann Arbor 0.73 88.12O 279 88,138 352 270 GiG Stage IV (0.54, 0.98) H PSMB8R141C: Ann Arbor 0.72 95,127 278 93.143 360 270 CFC Stage IV (0.54, 0.95) He PSMB8/W182M: Ann Arbor 0.72 95/127 278 93.143 360 270 GiG Stage IV (0.54, 0.95) He PSMB9, G9E: Ann Arbor O.71 95.126 278 93.142 358 270 GiG Stage IV (0.53, 0.94) He PSMB9, V32I: Ann Arbor O.69 93.123 277 89,138 363 270 CFC Stage IV (0.51, 0.93) He -T-I-T- O.O.5. O.1 O2 0.5 1

US 9,322,066 B2 67 68 APPENDIX 2, TABLE 2.7 Duration of response by protein expression and by covariate, IRC review. * (All reported groups are significant (ps 0.05) and with a frequency e10%)

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

CD68 OVERALL High 3.66 7.11 423 7/8 2O2 91 POSITIVE: O-25 FLIPI (1.10, 12.16) H Score

P279% NUCLEI High O.14 7/9 337 3f6 1162 96 POSITIVE: 60-70 FLIPI (0.03, 0.74) He Score

20S INTENSITY Intermediate 2.39 9.25 554 20,26 264 99 CYTOPLASMIC FLIPI (1.08, 5.26) H SIGNAL: s2- Score

20S INTENSITY Intermediate O.38 1725 352 10,23 813 99 CYTOPLASMIC FLIPI (0.17, 0.84) H SIGNAL: e3+ Score

2OSINTENSITY No Prior O.S3 22.32 427 20.43 638 155 CYTOPLASMIC Rituximab (0.29, 0.97) H SIGNAL: e3+ Therapy

2OS 9/o NUCLEAR Prior 2.39 9.13 356 13,14 245 97 STAINING: 30-50 Rituximab (0.97, 5.86) H Therapy

CD68 OVERALL s1 year 9.97 5/7 3S4 5/5 145 79 POSITIVE: O-25 since last (1.13, 88.13) H anti lymphoma treatinent

2OSINTENSITY sé5 years 0.55 28.39 417 20:44 654 177 CYTOPLASMIC old (0.31, 0.98) H SIGNAL: e3+

CD68 POSITIVE Female O.38 12.20 354 8:15 648 131 FOLLICULAR: (0.15, 1.01) H O-2S

CD68 OVERALL Male 0.27 5/7 279 612 784 98 POSITIVE: 51-75 (0.07, 1.03) H

CD68 OVERALL Male O.10 3.3 284 2/7 98 POSITIVE: >75 (0.01, 1.02) H

CD68 POSITIVE Male O.11 3.3 284 2/7 86 PERIFOLLICULAR: (0.01, 1.11) H >75

P279% NUCLE Male O.29 5.6 281 7/9 502 99 POSITIVE: 60-70 (0.08, 1.08) H

P65 INTENSITY Ann Arbor O16 2.4 173 5.8 474 87 CYTOPLASMIC Stage III (0.02, 1.19) H SIGNAL: s1

1 10 10 US 9,322,066 B2 69 70 APPENDIX 2, TABLE 2.8 Duration of Response by Germline Genetic Variant and by Covariate, IRC Review. (All Reported Groups are Significant (ps 1.05) and at a Frequency is 10%)

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

PSMB9,R6OH: Male O42 1011 211 19.28 351 120 AG (0.19, 0.94) H

PSMBSR24C: Ann Arbor 3.29 418 13,2O 438 43 CFC Stage II (1.07, 10.09) H

O.O1 O.O.5. O.1 O.S 1 5 10

APPENDIX 2, TABLE 2.9 Duration of Response by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (Ps 0.05) and at a Frequency s10%

Marker: Marker: HR HR R R Vc - R Vc - R Subgroup Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2. No O.15 4.5 343 9.15 488 177 P59L: Subgroup (0.03, 0.71) He MD

BCL2. 1 Prior O.11 2.2 345 5/10 488 82 P59L:MD TherapyLine of (0.01, 1.21) H

BCL2. Prior O.14 3.3 330 4.5 372 68 P59L:MD RituximabTherapy (0.01, 1.33) H

BCL2. >1 year O.26 5/5 356 48 502 115

P46L:MD sinceanti last (0.06, 1.09) H lymphoma treatinent BCL2. >1 year O.13 3f4 330 6.10 502 116 P59L: since last (0.02, 0.82) He MD anti lymphoma treatinent BCL2. sé5 years O.61 32.49 353 32,66 648 128 E29K: old (0.37, 1.00) H MIND

NOTCH1 s65 years O.66 44f64 369 4584. 490 152 X28INS: old (0.43, 1.00) H MIND

NOTCH/ >65 years 2.28 10.23 948 18,22 356 62 P251.3Li: old (1.02, 5.09) H MIND

BCL2. White O.10 4.5 343 814 488 163 P59L. (0.02, 0.57) He MD

BCL2 Ann Arbor 0.57 23:28 353 29,50 488 78 A43G: Stage IV (0.33, 1.00) H MIND US 9,322,066 B2 71 72 APPENDIX 2, TABLE 2.9-continued Duration of Response by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (Ps 0.05) and at a Frequency s10%

Marker: Marker: HR HR R R Vc - R Vc - R Subgroup Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

BCL2 Ann Arbor O.S1 2126 344 23:43 490 78 E29K: Stage IV (0.28, 0.93) H MIND

BCL2 Ann Arbor O.S4 21.26 352 27.48 423 90 R106H: Stage IV (0.30, 0.96) H MIND

NOTCH Ann Arbor O.S8 26,31 353 35.58 423 91 X28INS: Stage IV (0.35, 0.98) H MIND

O.OO5 O.O1 O.05 0.1 0.5 1 5 10

APPENDIX 2, TABLE 2.10 Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal CD68 OVERALL No 0.41 30.44 409 1941 104.7 442 POSITIVE: O-25 Subgroup (0.23, 0.73) He

CD68 POSITIVE No O.S4 40.60 462 29,51 834 387 FOLLICULAR: O-2S Subgroup (0.33, 0.88) H

CD68 POSITIVE No O.48 26,39 374 20:41 1103 384 PERIFOLLICULAR: O-2S Subgroup (0.27, 0.87) H

P27 SIGNAL No 0.77 127,199 533 114,196 700 463 TENSITY:2- Subgroup (0.60, 0.99) |-

P659. POSITIVE No 0.75 129,204 505 108,186 726 470 CYTOPLASMIC Subgroup (0.58, 0.96) |- SIGNAL: -90%

CD68 POSITIVE Prior O.43 15.18 437 1016 744 174 PERIFOLLICULAR: -75 Line o (0.19, 0.97) H Therapy

P279% NUCLEI Prior O.38 19:26 434 616 1047 204 POSITIVE: O-20 Line o (0.15, 0.97) H Therapy

P659% Prior O.63 54.79 S46 39,73 841 2O3 NUCLEAR Line o (0.41, 0.95) H STAINING: O Therapy

P659. POSITIVE Prior O.61 59.89 550 47,88 841 2O3 CYTOPLASMIC Line o (0.42, 0.90) H SIGNAL: -90% Therapy

P65 INTENSITY Prior O.64 54.85 568 47.91 841 2O3 CYTOPLASMIC Line o (0.43, 0.95) H SIGNAL:2- Therapy US 9,322,066 B2 73 74 APPENDIX 2, TABLE 2.10-continued Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Si ificant is 0.05) and at a Frequency 10%

Marker: HR HR R R Vc - R. Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal CD68 OVERALL O.20 6.8 204 2/7 1005 111 POSITIVE: O-25 (0.04, 1.04) H

CD68 POSITIVE O.09 6.8 171 1/7 98 PERIFOLLICULAR: O-2S (0.01, 0.81) H

P659% O.31 7/9 225 7,12 705 125 NUCLEAR (0.10, 0.92) STAINING:sso H

P279% NUCLEI 8.15 16 5/5 227 74 POSITIVE: 60-70 (0.93, 71.30) H

P27 SIGNAL O.20 5.6 228 3.8 939 16 INTENSITY:2- (0.04, 1.02) H

P659% O.13 3.3 235 2.4 1235 16 NUCLEAR (0.01, 1.34) STAINING:sso H

CD68 OVERALL O.23 814 552 5.19 1235 204 POSITIVE: O-25 (0.07, 0.79) H

CD68 OVERALL No High 2.27 10.25 12.18 511 204 POSITIVE: 51-75 O (0.98, 5.28) Burden

CD68 POSITIVE No High O.38 1422 573 1225 939 182 FOLLICULAR: O-2S O (0.17, 0.87) Burden

CD68 POSITIV E O.22 8.13 5O1 S$1.6 1235 182 PERIFOLLICULAR: O-2S (0.06, 0.74)

P279% NUCLE O.S4 23, 27 220 23.31 503 248 POSITIVE: O-20 O (0.30, 0.98) Burden

P279% NUCLE High O.45 1722 421 13.25 599 248 POSITIVE: 30-50 O (0.22, 0.93) Burden

P27 SIGNAL High O.42 1618 220 11.17 975 248 INTENSITY: s1 O (0.19, 0.93) Burden

CD68 OVERALL Intermediate 0.27 7.11 485 5/15 1235 159 POSITIVE: O-25 FLIPI (0.08, 0.96) Score

CD68 POSITIVE Intermediate 0.44 1420 421 11,2O 939 141 FOLLICULAR: O-2S FLIPI (0.19, 1.01) Score

CD68 POSITIVE Intermediate 0.27 713 5O1 5/17 1235 139 PERIFOLLICULAR: O-2S FLIPI (0.08, 0.95) Score

CD68 OVERALL Low O.32 8.10 422 5/11 1075 102 POSITIVE: O-25 FLIPI (0.10, 0.99) Score US 9,322,066 B2 75 76 APPENDIX 2, TABLE 2.10-continued Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

CD68 OVERALL No Prior O.36 12.19 533 7.21 1103 241 POSITIVE: O-25 Rituximab (0.14, 0.93) H Therapy

CD68 POSITIVE No Prior O.40 23,36 485 10.25 1103 210 FOLLICULAR: O-2S Rituximab (0.19, 0.84) H Therapy

CD68 POSITIVE No Prior O.33 12.18 533 5/15 208 PERIFOLLICULAR: O-2S Rituximab (0.11, 0.95) H Therapy

2OS 9/o NUCLEAR Prior O.39 1721 343 612 1075 212 STAINING: 60-70 Rituximab (0.15, 1.00) H Therapy

CD68 OVERALL Prior O41 18.25 235 12.20 834 2O1 POSITIVE: O-25 Rituximab (0.19, 0.89) H Therapy

P27 SIGNAL Prior O.64 60,88 421 48.85 718 210 INTENSITY:2- Rituximab (0.44, 0.94) H Therapy

P659% Prior O.29 1621 232 13,27 975 215 NUCLEAR Rituximab (0.14, 0.64) He STAINING:sso Therapy

CD68 OVERALL >1 year O.26 17.26 409 9.28 1235 269 POSITIVE: O-25 since las (0.11, 0.60) He anti lymphoma treatmen

CD68 POSITIVE >1 year O.48 22.34 550 1633 1005 235 FOLLICULAR: O-2S since las (0.25, 0.93) H anti lymphoma treatmen

CD68 POSITIVE >1 year O.40 1S,24 533 11.27 1235 234 PERIFOLLICULAR: O-2S since las (0.18, 0.88) H anti lymphoma treatmen

P659. POSITIVE >1 year O.68 69,117 593 S4f112 983 288 CYTOPLASMIC since las (0.48, 0.98) H SIGNAL: -90% anti lymphoma treatmen

2OSINTENSITY European O.S8 34,52 533 25,52 1075 214 CYTOPLASMIC Union (0.34, 0.97) H SIGNAL: e3+

CD68 OVERALL European O.40 1422 374 11.25 1103 211 POSITIVE: O-25 Union (0.18, 0.90) H

CD68 POSITIVE European O.35 13.17 675 7.14 1185 18S PERIFOLLICULAR: -75 Union (0.13, 0.96) H

P279% NUCLEI European O.S9 29,41 655 27/51 939 214 POSITIVE: 80-100 Union (0.35, 1.00) H US 9,322,066 B2 77 78 APPENDIX 2, TABLE 2.10-continued Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal P27 SIGNAL European 0.67 57,89 649 49,91 939 214 INTENSITY:2- Union (0.46,0.99) H

P659. POSITIVE European 0.67 57,92 552 48.87 975 216 CYTOPLASMIC Union (0.46,0.99) H SIGNAL: -90%

P65 INTENSITY European O.45 13,18 409 1121 1005 216 CYTOPLASMIC Union (0.19, 1.02) H SIGNAL: s1

CD68 OVERALL sé5 years O.38 2434 374 15.32 1005 329 POSITIVE: O-25 O (0.20, 0.73) H

CD68 POSITIVE sé5 years O.45 33,46 332 2441 764 292 FOLLICULAR: O-2S O (0.26, 0.77) H

CD68 POSITIVE sé5 years O.48 20.31 332 13,29 1075 289 PERIFOLLICULAR: O-2S O (0.24, 0.98) H

CD68 POSITIVE sé5 years O.45 1923 437 13,23 726 289 PERIFOLLICULAR: -75 O (0.22, 0.91) H

P279% NUCLE sé5 years O.S2 26/34 327 1932 834 344 POSITIVE: O-20 O (0.29, 0.96) H

P27 SIGNAL sé5 years O.74 103,153 484 90,149 672 344 INTENSITY:2- O (0.56, 0.98) |-

P659% sé5 years O.S4 1929 374 22.43 975 349 NUCLEAR O (0.29, 1.00) H STAINING:sso

P659. POSITIVE sé5 years O.71 103,154 484 84f142 719 349 CYTOPLASMIC O (0.53, 0.94) H SIGNAL: -90%

P65 INTENSITY sé5 years 0.72 101,152 484 82,139 718 349 CYTOPLASMIC O (0.54, 0.96) |- SIGNAL:2-

2OS 90 Female O.S3 3S,61 568 1949 259 NUCLEAR (0.30, 0.93) H STAINING: O-20

CD68 OVERALL Female 0.27 1626 434 S.18 1103 242 POSITIVE: O-25 (0.10, 0.76) He

CD68 POSITIVE Female O.45 23,37 462 10,22 1103 217 FOLLICULAR: O-2S (0.21, 0.96) H

CD68 POSITIVE Female O.29 19.28 374 5/17 1107 215 PERIFOLLICULAR: O-2S (0.11, 0.79) He

P65 INTENSITY Female O.68 74.129 537 41.87 1047 263 CYTOPLASMIC (0.46, 1.00) H SIGNAL:2- US 9,322,066 B2 79 80 APPENDIX 2, TABLE 2.10-continued Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Si ificant is 0.05) and at a Frequency 10%

Marker: HR HR Vc - R. Vc - R. Subgroup Marker: Level Subgroup (95% CI) (log scale) Median Evti N Median NTotal 2OS 90 Male O.42 1416 247 12.20 518 204 NUCLEAR (0.19, 0.93) STAINING: 60-70

2OSINTENSITY Male O.S8 2027 421 37.62 764 204 CYTOPLASMIC (0.34, 1.00) SIGNAL: e3+

CD68 OVERALL Male O46 1418 251 1423 975 200 POSITIVE: O-25 (0.22, 0.99)

CD68 OVERALL Male O.23 8.9 427 411 200 POSITIVE: >75 (0.07, 0.79)

P659. POSITIVE Male O.68 51/72 427 6297 726 CYTOPLASMIC (0.47, 0.98) SIGNAL: -90%

CD68 POSITIVE Asian O.13 5.6 220 2.6 30 FOLLICULAR: O-2S (0.01, 1.09)

CD68 POSITIVE Other O.14 44 341 4.5 658 22 PERIFOLLICULAR: 26-SO (0.01, 1.31)

CD68 OVERALL White O.38 24,35 409 16.35 1075 385 POSITIVE: O-25 (0.20, 0.72)

CD68 POSITIVE White 0.57 32.48 462 25.42 834 335 FOLLICULAR: O-2S (0.34, 0.98)

CD68 POSITIV E White O.43 1927 1939 1103 332 PERIFOLLICULAR: O-2S (0.22, 0.82)

P27 SIGNAL White 0.75 113,177 533 99.170 718 403 INTENSITY:2- (0.57, 0.98)

P659. POSITIVE White 0.73 113,179 505 95.162 744 406 CYTOPLASMIC (0.56, 0.96) SIGNAL: -90%

P65 INTENSITY White 0.75 109,176 533 744 406 CYTOPLASMIC (0.57, 0.99) SIGNAL:2-

CD68 OVERALL Ann Arbor O.35 1216 276 9,12 905 144 POSITIVE: O-25 Stage III (0.14, 0.92)

CD68 POSITIVE Ann Arbor O.40 1924 332 13,18 764 135 FOLLICULAR: O-2S Stage III (0.19, 0.83) H

P279% NUCLEI Ann Arbor O.42 1416 418 12.15 613 149 POSITIVE: O-20 Stage III (0.17, 1.00)

P65 INTENSITY Ann Arbor O.36 1011 261 7.11 834 151 CYTOPLASMIC Stage III (0.13, 0.99) SIGNAL: s1 . US 9,322,066 B2 81 82 APPENDIX 2, TABLE 2.10-continued Time to Next Anti-Lymphoma Therapy by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

CD68 OVERALL Ann Arbor 0.37 15,22 434 8.23 1047 222 POSITIVE: O-25 Stage IV (0.15, 0.92) H

CD68 POSITIVE Ann Arbor 0.37 11.15 374 8.22 1107 183 PERIFOLLICULAR: O-2S Stage IV (0.14, 0.96) H

O.O1 O.1 1 10

APPENDIX 2, TABLE 2.11 Time to Next Anti-Lymphoma Therapy by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal PSMB1, A171S: No O.78 173,276 S46 1S2,266 719 S42 Gif G Subgroup (0.63, 0.97) H

PSMB1 I2O8N: No O.78 173,276 S46 1S2,266 719 S42 TT Subgroup (0.63, 0.97) H

PSMB1AP11A: No O.68 78.127 550 63,115 939 S42 C.G. Subgroup (0.48, 0.94) H

PSMB1AP11A: No 0.57 3O37 436 26,43 613 S42 Gif G Subgroup (0.34, 0.97) H

PSMB1 P193Li: No O.78 73,276 546 S2,266 719 S42 CC Subgroup (0.63, 0.97) H

PSMB2 E49X: No O.78 73,276 546 S2,266 719 S42 Gif G Subgroup (0.63, 0.97) H

PSMB2 G187V: No O.78 73,276 546 S2,266 719 S42 Gif G Subgroup (0.63, 0.97) H

PSMB2FL159F: No O.78 73,276 546 S2,266 719 S42 CC Subgroup (0.63, 0.97) H

PSMBSFL2O6M: No O.78 73,276 546 S2,266 719 S42 CC Subgroup (0.63, 0.97) H

PSMBSR24C: No O.78 47,235 S46 27,223 719 S42 CC Subgroup (0.62, 0.99)

PSMB6 A234D: No O.78 73,276 546 S2,266 719 S42 CC Subgroup (0.63, 0.97) H

US 9,322,066 B2 85 86 APPENDIX 2, TABLE 2.11-continued Time to Next Anti-Lymphoma Therapy by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal PSMB8, G8R: 1 Prior O.68 69,109 673 55,111 852 23 Gif G Line of (0.47, 0.97) H Therapy

PSMB8. R141C: 1 Prior O.70 72,113 673 61.118 841 23 CC Line of (0.50, 0.98) H Therapy

PSMB8, W182M: 1 Prior O.70 72,113 673 61.118 841 23 Gif G Line of (0.50, 0.98) |- Therapy

PSMB9, G9E: 1 Prior O.71 71,112 673 61.118 841 23 Gif G Line of (0.50, 1.00) H Therapy

PSMB1, A171S: S Prior O.29 7/9 235 612 939 2 Gif G Lines of (0.08, 0.99) H Therapy

PSMB1 I2O8N: S Prior O.29 7/9 235 612 939 2 TT Lines of (0.08, 0.99) H Therapy

PSMB1 P193Li: S Prior O.29 7/9 235 612 939 2 CC TherapyLines of (0.08, 0.99) H

PSMB2 E49X: S Prior O.29 7/9 235 612 939 2 Gif G Lines of (0.08, 0.99) H Therapy

PSMB2 G187V: S Prior O.29 7/9 235 612 939 2 Gif G Lines of (0.08, 0.99) H Therapy

PSMB2FL159F: S Prior O.29 7/9 235 612 939 2 CC Lines of (0.08, 0.99) H Therapy

PSMBSFL2O6M: S Prior O.29 7/9 235 612 939 2 CC TherapyLines of (0.08, 0.99) H

PSMBSR24C: S Prior O16 6.7 235 4.9 1235 2 CC Lines o (0.03, 0.79) He Therapy

PSMB6 A234D: S Prior O.29 7/9 235 612 939 2 CC TherapyLines of (0.08, 0.99) H

PSMB6P107A: S Prior O.29 7/9 235 612 939 2 CC Lines of (0.08, 0.99) H Therapy

PSMB8, G8R: S Prior O.29 7/9 235 611 939 2 Gif G Lines of (0.08, 1.00) H Therapy

PSMB8. R141C: S Prior O.29 7/9 235 612 939 2 CC Lines of (0.08, 0.99) H Therapy

PSMB8, W182M. S. Prior O.29 7/9 235 612 939 2 Gif G TherapyLines of (0.08, 0.99) H

US 9,322,066 B2 89 90 APPENDIX 2, TABLE 2.11-continued Time to Next Anti-Lymphoma Therapy by Germline Genetic Variant and by Covariate, IRC Review. (All Reported Groups are Significant (ps 0.05) and at a Frequency of >10%)

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

PSMB9 V32I: High O.71 106.142 396 94f142 S21 296 CC Tumor (0.54, 0.94) H Burden

PSMB1, A171S: >1 year O.69 |- 98.165 688 80,166 969 331 Gif G since last (0.51, 0.93) anti lymphoma treatmen

PSMB1 I2O8N: >1 year O.69 98.165 688 80,166 969 331 TT since last (0.51, 0.93) |- anti lymphoma treatmen

PSMB1AP11A: >1 year O.47 1720 429 13.25 872 331 Gif G since last (0.23, 0.97) H anti lymphoma treatmen

PSMB1 P193Li: >1 year O.69 98.165 688 80,166 969 331 CC since last (0.51, 0.93) |- anti lymphoma treatmen

PSMB2 E49X: >1 year O.69 98.165 688 80,166 969 331 Gif G since last (0.51, 0.93) |- anti lymphoma treatmen

PSMB2/G187V: >1 year O.69 98.165 688 80,166 969 331 Gif G since last (0.51, 0.93) |- anti lymphoma treatmen

PSMB2FL159F: >1 year O.69 98.165 688 80,166 969 331 CC since last (0.51, 0.93) |- anti lymphoma treatmen

PSMB5/L206M; >1 year O.69 98.165 688 80,166 969 331 CC since last (0.51, 0.93) |- anti lymphoma treatmen

PSMBSR24C: >1 year O.70 81,137 649 68,141 983 331 CC since last (0.51, 0.97) |- anti lymphoma treatmen

PSMB6 A234D: >1 year O.69 |- 98.165 688 80,166 969 331 CC since last (0.51, 0.93) anti lymphoma treatmen

PSMB6/P107A: >1 year O.70 95,159 702 79,162 969 331 CC since last (0.52, 0.94) H anti lymphoma treatmen US 9,322,066 B2 91 92 APPENDIX 2, TABLE 2.11-continued Time to Next Anti-Lymphoma Therapy by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal PSMB8, G8R: >1 year O.69 91.155 702 74,159 1005 331 Gif G since last (0.51, 0.93) H anti lymphoma treatmen PSMB8/R141C: >1 year O.69 98.165 688 80,166 969 331 CC since last (0.51, 0.93) H anti lymphoma treatmen PSMB8/V182M: >1 year O.69 98.165 688 80,166 969 331 Gif G since last (0.51, 0.93) H anti lymphoma treatmen PSMB9, G9E: >1 year O.69 97.163 688 80.16S 969 331 Gif G since last (0.51, 0.93) H anti lymphoma treatmen PSMB9 V32I: >1 year O.68 96.160 649 76.158 969 331 CC since last (0.50, 0.91) H anti lymphoma treatmen PSMB1A171S: European 0.71 75,119 675 66.122 939 24 Gif G Union (0.51, 0.99) |-

PSMB1 I2O8N: European O.71 75,119 675 66,122 939 24 TT Union (0.51, 0.99) |-

PSMB1 P11A: European O.S6 34.f49 462 29,51 1OOS 24 C.G. Union (0.34, 0.92) H

PSMB1 P193L: European 0.7 75,119 675 66,122 939 24 CC Union (0.51, 0.99) |-

PSMB2E49X: European 0.7 75,119 675 66,122 939 24 Gif G Union (0.51, 0.99) |-

PSMB2G187V: European 0.7 75,119 675 66,122 939 24 Gif G Union (0.51, 0.99) |-

PSMB2L159F: European 0.7 75,119 675 66,122 939 24 CC Union (0.51, 0.99) |-

PSMBSL206M: European 0.7 75,119 675 66,122 939 24 CC Union (0.51, 0.99) |-

PSMB6A234D: European 0.7 75,119 675 66,122 939 24 CC Union (0.51, 0.99) -

PSMB8, G8R: European 0.7 70,111 649 62,117 939 24 Gif G Union (0.50, 1.00) |-

US 9,322,066 B2 97 98 APPENDIX 2, TABLE 2.11-continued Time to Next Anti-Lymphoma Therapy by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

PSMBSR24C: Ann O45 11.15 421 13,28 991 270 CT Arbor (0.20, 1.01) H Stage IV

PSMB6 A234D: Ann 0.72 86.127 457 78.143 639 270 CC Arbor (0.53, 0.97) |- Stage IV

PSMB6P107A: Ann 0.72 84f124 485 77,140 602 270 CC Arbor (0.53, 0.98) |- Stage IV

PSMB8, G8R: Ann 0.73 80.12O 457 74,138. 602 270 Gif G Arbor (0.53, 1.00) |- Stage IV

PSMB8. R141C: Ann 0.72 86.127 457 78.143 639 270 CC Arbor (0.53, 0.97) |- Stage IV

PSMB8, W182M: Ann 0.72 86.127 457 78.143 639 270 Gif G Arbor (0.53, 0.97) |- Stage IV

PSMB9, G9E: Ann O.71 86.126 457 78,142 602 270 Gif G Arbor (0.52, 0.97) |- Stage IV

PSMB9 V32I: Ann 0.72 84,123 457 76.138 639 270 CC Arbor (0.53, 0.98) - Stage IV

O.O1 O.1 1 10 1OO

APPENDIX 2, TABLE 2.12 Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2A43G: No O.66 104,161 5O1 84f154 791 315 MIND Subgroup (0.49, 0.88) He

BCL2. No O.71 127,192 SO1 102.179 719 390 C.-11C > T: Subgroup (0.54, 0.92) He MIND

BCL2, E29K: No O.64 98.1SO 484 77,139 791 3.18 MIND Subgroup (0.47, 0.86) He

BCL2 P46L: No O.70 86.13S SO4 77,139 743 314 MIND Subgroup (0.51, 0.95) He

BCL2.P46S: No 0.73 92,147 SO4 84f147 751 310 MIND Subgroup (0.54, 0.98) He US 9,322,066 B2 99 100 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2fPS9S: No 0.72 92,145 SO4 84f147 751 309 MIND Subgroup (0.53, 0.96) He

BCL2/Q52P: No O.71 97,153 501 88,154. 751 307 MIND Subgroup (0.53, 0.95) He

BCL2iR106H: No O.66 O8,162 489 84f151. 751 372 MIND Subgroup (0.50, 0.88) He

NOTCHA No 0.73 13,175 SO1 93,161 726 337 G A1702P: Subgroup (0.55, 0.96) He MIND

NOTCHAI1681N: No 0.73 13,175 SO1 94f162 719 337 MIND Subgroup (0.56, 0.97) He

NOTCHAL1679P: No 0.73 13,175 SO1 94f162 719 337 MIND Subgroup (0.56, 0.97) He

NOTCH/L1679Q: No 0.73 13,175 SO1 94f162 719 337 MIND Subgroup (0.56, 0.97) He

NOTCHL2458V: No 0.75 104,162 501 92.157 719 319 MIND Subgroup (0.56, 0.99) He

NOTCHP251.3Li: No O.69 90,140 SO1 67,123 791 361 MIND Subgroup (0.50, 0.95) He

NOTCHP2515FS4: No O.70 114,176 SO1 92.166 751 344 MIND Subgroup (0.53, 0.92) He

NOTCH/Q2441X: No 0.75 103,161 SO1 92.15S 719 319 MIND Subgroup (0.57, 1.00) He

NOTCH/Q246OX: No 0.72 98.151. 489 82,143 726 324 MIND Subgroup (0.53, 0.96) He

NOTCHSX26DEL: No O.74 112,175 SOS 102,178 744 361 MIND Subgroup (0.57, 0.97) He

NOTCHSX26INS: No O.74 114,179 SOS 103,180 744 361 MIND Subgroup (0.56, 0.96) He

NOTCHSX28DEL: No 0.72 128,194 SO3 109,189 719 400 MIND Subgroup (0.56, 0.93) He

NOTCHSX28INS: No O.70 132,198 SO1 112.195 719 401 MIND Subgroup (0.55, 0.90) He

BCL2 A43G: MND 1 Prior O.S4 46.72 SSO 30.67 991 139 Line of (0.34, 0.86) He Therapy US 9,322,066 B2 101 102 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2.C.-11C > T: Prior O.61 57,86 568 38.78 988 70 MIND Line o (0.40, 0.92) He Therapy

BCL2 E29K: MIND Prior O.S4 4164 S46 29.62 991 39 Line o (0.34, 0.88) He Therapy

BCL2 P46L: MND Prior O.S9 37/58 589 30.63 991 40 Line o (0.37, 0.97) He Therapy

BCL2 P46S: MIND Prior O.61 40.64 550 33.68 991 38 Line o (0.38, 0.97) He Therapy

BCL2 PS9S: MIND Prior O.62 41.66 573 32,66 988 38 Line o (0.39, 0.99) He Therapy

BCL2/Q52P: MND Prior O.62 42.67 550 35.71 988 38 Line o (0.39, 0.97) He Therapy

BCL2 R106H: MIND Prior O.S8 49,73 550 33.68 991 62 Line o (0.37, 0.90) He Therapy

NOTCHP2515FS4: 1 Prior 0.62 48,75 573 35/72 988 149 MIND Line o (0.40, 0.97) He Therapy

NOTCHSX28DEL: 1 Prior 0.67 56/85 568 4484 756 177 MIND Line o (0.45, 1.00) He Therapy

NOTCHSX28INS: 1 Prior O.63 56.84 568 45.88 841 178 MIND Line o (0.42, 0.93) He Therapy

BCL2.C.-11C > T: S Prior O.17 3.3 221 3/7 939 11 MIND Lines of (0.03, 1.05) He Therapy

NOTCHSX28DEL: S Prior O.20 44 342 3/7 939 11 MIND Lines of (0.04, 1.12) He Therapy

BCL2 A43G: MND No High O.61 42.78 774 27.64 42 Tumor (0.37, 0.98) He Burden

BCL2 R106H: MIND No High O.60 46.79 690 26.58 118S 63 Tumor (0.37, 0.98) He Burden

NOTCHP251.3Li: No High 0.44 17.26 593 9,21 63 MD Tumor (0.19, 0.99) He Burden

NOTCHSX28INS: No High O.63 55.98 734 37.82 1107 84 MIND Tumor (0.41, 0.96) He Burden

BCL2 A43G: MND High O.60 62.83 380 57.90 534 73 Tumor (0.42, 0.86) He Burden US 9,322,066 B2 103 104 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2.C.-11C > T: High O.65 76.101 380 69,107 484 217 MIND Tumor (0.47, 0.90) He Burden

BCL2 E29K: MIND High O.S6 60,79 374 52.82 S42 74 Tumor (0.38, 0.81) He Burden

BCL2 P46L: MND High 0.57 51,67 374 51.83 534 73 Tumor (0.38, 0.84) He Burden

BCL2 P46S: MIND High O.63 52/73 374 S6,89 488 70 Tumor (0.43, 0.92) He Burden

BCL2 PS9L: MND High O.66 52/72 396 SO.80 455 72 TumorBurden (0.44, 0.97) He

BCL2 PS9S: MIND High O.62 53,73 396 54.87 534 70 Tumor (0.42, 0.90) He Burden

BCL2/Q52P: MND High O.62 55/76 374 S8.92 534 68 TumorBurden (0.42, 0.89) He

BCL2 R106H: MIND High O.S9 62.83 380 58.93 S42 209 Tumor (0.41, 0.84) He Burden

NOTCHF1593S: High O.69 55.77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCHAG A1702P: High O.62 67.90 374 62.96 534 87 MIND Tumor (0.44, 0.88) He Burden

NOTCHAI1681N: High O.63 67.90 374 63.97 534 87 MIND TumorBurden (0.45, 0.89) He

NOTCHAL1586P: High O.69 55.77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCH/L1586Q: High O.65 56,78 396 60.94 534 72 MIND Tumor (0.45, 0.94) He Burden

NOTCHAL1594P: High O.69 55.77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCHAL1597H: High O.65 56,78 396 6094 534 72 MIND Tumor (0.45, 0.94) He Burden

NOTCHA High O.69 55.77 410 60.91 488 68 L1597 S1598INSG: Tumor (0.48, 1.00) MIND Burden He

NOTCHAL16O1P: High O.68 55.77 410 59,90 488 68 MIND Tumor (0.47, 0.98) He Burden US 9,322,066 B2 105 106 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal NOTCHAL1679P: High O.63 67.90 374 63.97 S34 87 MIND Tumor (0.45, 0.89) He Burden

NOTCH/L1679Q: High O.63 67.90 374 63.97 S34 87 MIND Tumor (0.45, 0.89) He Burden

NOTCHAL2458W: High O.64 61,83 380 61.9S S34 78 MIND Tumor (0.45, 0.91) He Burden

NOTCHP251.3Li: High O.S4 S8.76 380 43.71 S.42 98 MIND Tumor (0.36, 0.80) He Burden

NOTCHP2515FS4: High O.61 67.90 380 61.98 S34 89 MIND Tumor (0.43, 0.87) He Burden

NOTCH/Q2441X: High O.64 61,83 380 61.9S S34 78 MIND Tumor (0.45, 0.91) He Burden

NOTCH/Q246OX: High O.61 59,79 380 54.87 542 81 MIND Tumor (0.42, 0.88) He Burden

NOTCH/R1599 > QS: High O.69 SS, 77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCHAR1599P: High O.65 56,78 396 60.94 534 72 MIND Tumor (0.45, 0.94) He Burden

NOTCHV1579DEL: High O.69 SS, 77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCHV1579E: High O.65 56,78 396 60.94 534 72 MIND Tumor (0.45, 0.94) He Burden

NOTCHV1579G: High O.69 SS, 77 410 60.91 488 68 MIND Tumor (0.48, 1.00) He Burden

NOTCHSX26DEL: High 0.67 63.88 396 68,107 S18 98 MIND Tumor (0.47, 0.94) He Burden

NOTCHSX26INS: High O.66 64,89 396 68,107 S18 98 MIND Tumor (0.47, 0.93) He Burden

NOTCHSX28DEL: High O.64 74, 99 374 72.1.11 488 216 MIND Tumor (0.46, 0.89) He Burden

NOTCHSX28INS: High O.64 77,100 343 75,113 474 217 MIND Tumor (0.47, 0.88) He Burden

BCL2 PS9L: MD High O.18 4f6 211 67 658 125 FLIPI (0.03, 1.06) He Score US 9,322,066 B2 107 108 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2 A43G: MND intermediate O.60 34,53 SO4 26,52 939 05 FLIP (0.36, 1.00) He Score

BCL2.C.-11C > T: intermediate O.S9 44f66 SO4 33.63 939 36 MIND FLIP (0.38, 0.93) He Score

BCL2 P46S: MIND intermediate O.S9 31.50 505 26,52 988 05 FLIP (0.35, 1.00) He Score

BCL2fPS9S: MIND intermediate O.60 31f48 505 27/52 939 05 ScoreFLIP (0.35, 1.00) He

BCL2/Q52P: MND intermediate O.60 32.51 505 28,54 988 05 FLIP (0.36,0.99) He Score

BCL2R106H: MND Intermediate O.S1 4OSS SO1 28,54 988 30 FLIP (0.31, 0.82) He Score

NOTCHX28DEL: Intermediate O.65 44f65 505 38,7O 756 43 MIND FLIP (0.42, 1.00) He Score

BCL2, E29K: MND No Prior O.65 57,93 534 41;76 883 86 Rituximab (0.43, 0.97) He Therapy

BCL2 PS9L: MD No Prior O.15 3.5 211 713 947 78 Rituximab (0.03, 0.80) He Therapy

BCL2 A43G: MND Prior O.S9 44f63 428 37,69 719 32 Rituximab (0.38, 0.92) He Therapy

BCL2.C.-11C > T: Prior O.S8 57/78 396 43.77 718 65 MIND Rituximab (0.39, 0.86) He Therapy

BCL2 E29K: MIND Prior O.S9 41 S7 428 36.63 719 32 Rituximab (0.37, 0.92) He Therapy

BCL2fPS9S: MIND Prior O.60 39;S6 450 37,69 719 33 Rituximab (0.38, 0.95) He Therapy

BCL2 R106H: MIND Prior O.S4 50.67 421 39.72 719 59 Rituximab (0.36, 0.83) He Therapy

NOTCHAG A1702P: Prior O.60 51/72 434 4174 719 47 MIND Rituximab (0.40, 0.91) He Therapy

NOTCHAI1681N: Prior O.61 51/72 434 42.75 719 47 MIND Rituximab (0.41, 0.93) He Therapy

NOTCH/L1586Q: Prior O.63 44f63 443 41.73 719 36 MIND Rituximab (0.41, 0.97) He Therapy US 9,322,066 B2 109 110 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of a 10%

Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

NOTCHAL1597H: Prior O.63 44f63 443 41.73 719 36 MIND Rituximab (0.41, 0.97) He Therapy

NOTCHAL1679P: Prior O.61 51/72 434 42.75 719 47 MIND Rituximab (0.41, 0.93) He Therapy

NOTCH/L1679Q: Prior O.61 51/72 434 42.75 719 47 MIND Rituximab (0.41, 0.93) He Therapy

NOTCHAL2458W: Prior O.61 47.66 443 42.75 719 41 MIND Rituximab (0.40, 0.93) He Therapy

NOTCHP251.3Li: Prior 0.57 39.57 450 25,51 751 S4 MIND Rituximab (0.35, 0.95) He Therapy

NOTCHP2515FS4: Prior O.61 51/72 434 4277 719 49 MIND Rituximab (0.41, 0.92) He Therapy

NOTCH/Q2441X: Prior O.61 47.66 443 42.73 719 41 MIND Rituximab (0.40, 0.93) He Therapy

NOTCH/Q246OX: Prior O.S8 45.62 434 36.66 719 43 MIND Rituximab (0.37, 0.90) He Therapy

NOTCHAR1599P: Prior O.63 44f63 443 41.73 719 36 MIND Rituximab (0.41, 0.97) He Therapy

NOTCHV1579E: Prior O.63 44f63 443 41.73 719 36 MIND Rituximab (0.41, 0.97) He Therapy

NOTCHSX26DEL: Prior O.65 50, 70 434 4682 718 58 MIND Rituximab (0.44, 0.98) He Therapy

NOTCHSX26INS: Prior O.64 52.74 434 47.84 719 58 MIND Rituximab (0.43, 0.94) He Therapy

NOTCHSX28DEL: Prior O.61 60,83 427 49.86 672 75 MIND Rituximab (0.42, 0.89) He Therapy

NOTCHSX28INS: Prior O.60 62.86 427 SO.88 672 75 MIND Rituximab (0.41, 0.87) He Therapy

BCL2 A43G: MND >1 year O.63 57.91 589 4494 878 85 since las (0.43, 0.94) He anti lymphoma treatmen

BCL2.C.-11C > T: >1 year O.62 73,112 S60 55,112 878 233 MIND since last (0.44, 0.88) He anti lymphoma treatmen US 9,322,066 B2 111 112 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median N Total BCL2 E29K: MIND >1 year O.S9 S2,82 550 39.83 883 18S since las (0.39, 0.90) anti He lymphoma treatmen

BCL2 R106H: MIND >1 year 0.57 65.96 534 45.94 878 217 since las (0.39, 0.83) anti He lymphoma treatmen

NOTCHP251.3Li: >1 year O.60 50,78 S60 37.81 878 213 MIND since las (0.39, 0.92) anti lymphoma treatmen

NOTCHP2515FS4: >1 year O.65 64,102 593 48,102 878 205 MIND since las (0.44, 0.94) anti lymphoma treatmen NOTCH/Q246OX: >1 year O.66 53.84 573 43.86 841 18S MIND since las (0.44, 0.99) anti lymphoma treatmen

NOTCHX26INS: >1 year O.69 62.102 643 54,110 883 212 MIND since las (0.48, 1.00) anti He lymphoma treatmen

NOTCHSX28DEL: >1 year O.70 70,112 S89 58,116 872 236 MIND since las (0.49, 0.99) anti He lymphoma treatmen

NOTCHSX28INS: >1 year O.64 73.113 573 59,119 872 237 MIND since las (0.45, 0.90) anti lymphoma reatmen

BCL2 A43G: MND European O.S3 41.66 649 32.72 1103 38 Union (0.33, 0.84)

BCL2.C.-11C > T: European 0.57 53.79 505 41.83 983 69 MIND Union (0.38, 0.86)

BCL2 E29K: MIND European O.S8 37,61 533 3267 1075 38 Union (0.36, 0.94)

BCL2 P46L: MND European O.S8 34,56 533 2964 118S 39 Union (0.35, 0.96)

BCL2 P46S: MIND European O.S9 37,61 SO4 33.68 1075 36 Union (0.37, 0.94)

BCL2fPS9S: MIND European O.60 36.60 505 34f69 983 36 Union (0.37, 0.96) US 9,322,066 B2 113 114 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal BCL2/Q52P: MND European O.60 38.63 SO4 36,73 983 36 Union (0.38, 0.94) He

BCL2R106H: MND European 0.57 44f6S S33 35.70 100S 60 Union (0.36, 0.89) He

NOTCHG A1702P: European O.61 44f72 S33 36,74 983 47 MIND Union (0.39, 0.95) He

NOTCHAI1681N: European O.63 44f72 533 37.75 983 47 MIND Union (0.40, 0.97) He

NOTCH/L1586Q: European O.63 40,66 SOS 36.72 983 38 MIND Union (0.40, 0.99) He

NOTCHAL1597H: European O.63 40,66 SOS 36.72 983 38 MIND Union (0.40, 0.99) He

NOTCHAL1679P: European O.63 44f72 533 37.75 983 47 MIND Union (0.40, 0.97) He

NOTCH/L1679Q: European O.63 44f72 533 37.75 983 47 MIND Union (0.40, 0.97) He

NOTCHP251.3Li: European O.47 39,59 SO4 22.52 59 MIND Union (0.28, 0.80) He

NOTCHAP2S1SFS4: European O.S8 48,77 533 3677 1075 55 MIND Union (0.38, 0.90) He

NOTCHAR1599P: European O.63 40,66 SOS 36.72 983 38 MIND Union (0.40, 0.99) He

NOTCHV1579E: European O.63 40,66 SOS 36.72 983 38 MIND Union (0.40, 0.99) He

NOTCHSX26DEL: European O.64 44f72 SS2 42.84 100S 59 MIND Union (0.42, 0.99) He

NOTCHSX26INS: European O.63 46.74 SS2 42.84 100S 59 MIND Union (0.42, 0.96) He

NOTCHSX28DEL: European O.62 53.82 533 45.87 983 77 MIND Union (0.41, 0.92) He

NOTCHSX28INS: European O.61 S6,8S S33 46,89 983 77 MIND Union (0.41, 0.90) He

BCL2/A43G: MND sé5 years O.66 84f1.24 484 65.11S 743 239 old (0.48, 0.91) He

US 9,322,066 B2 117 118 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal NOTCHSX26INS: sé5 years O.70 93.138 5O1 77.133 726 273 MIND old (0.52, 0.95) He

NOTCH/X28DEL: sé5 years O.68 104,149 462 83.140 719 298 MIND old (0.51, 0.90) He

NOTCHSX28INS: sé5 years 0.67 106.149 462 85.143 718 298 MIND old (0.50, 0.89) He

BCL2 P46L: MND Female O.61 51.86 S3S 31.68 118S 18O (0.39, 0.96) He

BCL2fPS9S: MIND Female O.65 57.97 533 36,73 947 178 (0.43, 1.00) He

BCL2 R106H: MIND Female O.65 58/96 534 35,70 883 2O3 (0.42, 0.99) He

BCL2 A43G: MND Male 0.55 45.59 343 47.76 743 35 (0.36, 0.82) He

BCL2C-11C > T: Male O.62 57f77 380 58.93 651 82 MIND (0.43, 0.90) He

BCL2 E29K: MIND Male O.S1 43 SS 337 43.68 743 38 (0.33, 0.77) He

BCL2 P46L: MD Male O.11 44 382 6.10 878 34 (0.02, 0.63) He

BCL2 P46S: MIND Male O.64 38.52 396 49.75 695 33 (0.42, 0.99) He

BCL2/Q52P: MND Male O.62 39.53 396 50,78 695 31 (0.41, 0.95) He

BCL2 R106H: MIND Male O.61 SOf 66 396 49,81 695 69 (0.41, 0.90) He

NOTCHAG A1702P: Male O.60 47.63 337 S1,81 726 45 MIND (0.40, 0.89) He

NOTCHAI1681N: Male O.61 47 63 337 5282 695 45 MIND (0.41, 0.90) He

NOTCH/L1586Q: Male O.65 41 S6 396 S1.79 695 35 MIND (0.43, 0.98) He

NOTCHAL1597H: Male O.65 41 S6 396 S1.79 695 35 MIND (0.43, 0.98) He US 9,322,066 B2 119 120 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal NOTCHAL1679P: Male O.61 47 63 337 5282 695 45 MIND (0.41, 0.90) He

NOTCH/L1679Q: Male O.61 47 63 337 5282 695 45 MIND (0.41, 0.90) He

NOTCHAL2458W: Male O.60 44,59 337 S1.80 695 39 MIND (0.40, 0.91) He

NOTCHP251.3Li: Male O.S2 41 S4 343 37.63 791 55 MIND (0.33,0.82) He

NOTCHP2515FS4: Male 0.57 47.62 380 S1.84 743 48 MIND (0.38, 0.85) He

NOTCH/Q2441X: Male O.61 44,59 337 51.79 695 39 MIND (0.40, 0.91) He

NOTCH/Q246OX: Male O.14 3.3 127 8.12 645 42 MD (0.03, 0.71) He

NOTCH/Q246OX: Male O.60 42.57 343 43.70 726 42 MIND (0.39, 0.93) He

NOTCHAR1599P: Male O.65 41 S6 396 S1,79 695 35 MIND (0.43, 0.98) He

NOTCHV1579E: Male O.65 41 S6 396 S1,79 695 35 MIND (0.43, 0.98) He

NOTCHSX26DEL: Male O.S8 49.6S 396 S8.95 726 62 MIND (0.39, 0.85) He

NOTCHSX26INS: Male O.S6 SOf 66 396 S8.95 743 62 MIND (0.38, 0.82) He

NOTCHSX28DEL: Male O.63 56.75 396 61.97 695 8O MIND (0.44, 0.91) He

NOTCHSX28INS: Male O.62 59.78 396 63,100 651 81 MIND (0.44, 0.89) He

BCL2 A43G: MND White 0.67 89.140 SO4 76,139 834 279 (0.49, 0.91) He

BCL2.C.-11C > T: White O.70 110,168 SO4 90,158 726 342 MIND (0.53, 0.93) He

BCL2 E29K: MIND White O.63 83,129 48S 70,127 834 282 (0.46, 0.87) He US 9,322,066 B2 121 122 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

BCL2 P46L: MND White O.68 74/117 505 68,125 791 279 (0.49, 0.95) He

BCL2 P46S: MIND White O.70 81,130 SOS 74,131 834 275 (0.51, 0.96) He

BCL2 PS9L: MND White 0.72 80,129 S33 69,123 791 278 (0.52, 0.99) He

BCL2 PS9S: MIND White O.68 81,128 SOS 73,130 834 274 (0.50, 0.94) He

BCL2/Q52P: MND White O.69 85/135 504 77,137 834 272 (0.50, 0.94) He

BCL2 R106H: MIND White O.68 93.140 5O1 77,135 743 327 (0.50, 0.92) He

NOTCHAG A1702P: White 0.72 97,152 SO4 82,143 743 295 MIND (0.54, 0.97) He

NOTCHAI1681N: White 0.72 97,152 SO4 82,143 743 295 MIND (0.54, 0.97) He

NOTCHAL1679P: White 0.72 97,152 SO4 82,143 743 295 MIND (0.54, 0.97) He

NOTCH/L1679Q: White 0.72 97,152 SO4 82,143 743 295 MIND (0.54, 0.97) He

NOTCHAL2458W: White 0.73 89.140 SO4 81,139 743 279 MIND (0.54, 0.99) He

NOTCHP251.3Li: White 0.67 78.122 SO4 S9,108 841 319 MIND (0.48, 0.94) He

NOTCHP2515FS4: White O.68 99.154 SOS 81,148 834 304 MIND (0.50, 0.91) He

NOTCH/Q2441X: White O.74 88,139 SO1 81,137 726 279 MIND (0.54, 1.00) He

NOTCH/Q246OX: White O.70 85/131 501 73,127 743 284 MIND (0.51, 0.96) He

NOTCHSX26DEL: White 0.73 99.155 505 91.158 751 321 MIND (0.55, 0.97) He

NOTCHSX26INS: White 0.72 101,159 S33 92.16O 751 321 MIND (0.54, 0.96) He US 9,322,066 B2 123 124 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

NOTCHSX28DEL: White 0.73 111,169 SO4 98.167 726 351 MIND (0.55, 0.95) He

NOTCHSX28INS: White O.70 115,174 SO4 99.171 743 352 MIND (0.53, 0.92) He

BCL2 A43G: MND Ann O.60 48,71 450 42.83 658 S4 Arbor (0.40, 0.91) He Stage IV

BCL2.C.-11C > T: Ann O.64 60.84 450 S4f1 OO 639 93 MIND Arbor (0.44, 0.93) He Stage IV

BCL2 E29K: MIND Ann O.S9 46.67 450 37.73 651 S4 Arbor (0.38, 0.91) He Stage IV

BCL2 P46L: MND Ann O.62 35, S2 449 38.74 651 48 Arbor (0.39, 0.99) He Stage IV

BCL2 PS9L: MD Ann O.17 3.5 182 8, 12 725 49 Arbor (0.03, 0.88) He Stage IV

BCL2 R106H: MIND Ann O.S9 SO,72 449 42.83 651 86 Arbor (0.39, 0.89) He Stage IV

NOTCHAG A1702P: Ann O.61 S2A75 428 47.87 649 63 MIND Arbor (0.41, 0.90) He Stage IV

NOTCHAI1681N: Ann O.62 S2A75 428 48.88 649 63 MIND Arbor (0.42, 0.92) He Stage IV

NOTCHAL1679P: Ann O.62 S2A75 428 48.88 649 63 MIND Arbor (0.42, 0.92) He Stage IV

NOTCH/L1679Q: Ann O.62 S2A75 428 48.88 649 63 MIND Arbor (0.42, 0.92) He Stage IV

NOTCHAL2458W: Ann O.63 4668 434 46.86 651 S4 MIND Arbor (0.42, 0.95) He Stage IV

NOTCHP251.3Li: Ann 0.55 4261 449 34f69 791 76 MIND Arbor (0.35, 0.87) He Stage IV

NOTCHP2515FS4: Ann O.S9 S1,74 434 46.91 658 67 MIND Arbor (0.39, 0.87) He Stage IV

NOTCH/Q2441X: Ann O.63 4668 434 46.85 651 S4 MIND Arbor (0.42, 0.95) He Stage IV

NOTCH/Q246OX: Ann O.61 43.64 449 41,79 658 56 MIND Arbor (0.39, 0.93) He Stage IV US 9,322,066 B2 125 126 APPENDIX 2, TABLE 2.12-continued Time to Next Anti-Lymphoma Therapy by Somatic Mutation and by Covariate, IRC review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: HR HR R R Vc - R Vc - R Subgroup Marker: Level Subgroup (95% CI) (log scale) Evti N Median Evti N Median NTotal

NOTCHSX26DEL: Ann O.63 53.76 457 S4f101 651 18O MIND Arbor (0.43, 0.93) He Stage IV

NOTCHSX26INS: Ann O.63 54f79 450 S4f101 651 18O MIND Arbor (0.43, 0.91) He Stage IV

NOTCHSX28DEL: Ann O.63 61.85 449 S6,103 644 195 MIND Arbor (0.44, 0.91) He Stage IV

NOTCHSX28INS: Ann O.60 64;87 434 S71 OS 644 196 MIND Arbor (0.42, 0.86) He Stage IV | | | | | |

APPENDIX 2, TABLE 2.13 Treatment Free Interval by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total CD68 OVERALL No O.42 30.44 269 19:41 883 442 POSITIVE: O-25 Subgroup (0.23, He 0.75) CD68 POSITIVE No O.S6 40.60 307 29,51 671 387 FOLLICULAR: Subgroup (0.34, He O-2S 0.90) CD68 POSITIVE No O.49 26,39 234 20:41 924 384 PERIFOLLICULAR: Subgroup (0.27, He O-2S 0.89) P659. POSITIVE No 0.77 129,204 363 108,186 581 470 CYTOPLASMIC Subgroup (0.60, k SIGNAL: -90% 1.00) CD68 POSITIVE Prior O.45 15.18 297 1016 587 174 PERIFOLLICULAR: Line O (0.20, He >75 Therapy 1.01) P279% NUCLEI Prior O.40 19:26 293 6, 16 883 204 POSITIVE: O-20 Line C (0.16, He Therapy 1.01) P65% NUCLEAR Prior O.66 54f79 406 39,73 673 2O3 STAINING: O Line o (0.43, Hs Therapy 0.99) P659. POSITIVE Prior O.63 59.89 409 47,88 673 2O3 CYTOPLASMIC Line o (0.43, He SIGNAL: -90% Therapy 0.93) P65 INTENSITY Prior O.66 54,85 427 47.91 673 2O3 CYTOPLASMIC Line o (0.45, s SIGNAL:2- Therapy 0.98) CD68 OVERALL 2 Prior O.20 6.8 8O 2/7 844 111 POSITIVE: O-25 Lines of (0.04, Therapy 1.02) US 9,322,066 B2 127 128 APPENDIX 2, TABLE 2.13-continued Treatment Free Interval by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total CD68 POSITIVE 2 Prior O.09 6.8 64 1/7 98 PERIFOLLICULAR: Lines o (0.01, He O-2S Therapy 0.81) P65% NUCLEAR 2 Prior O.32 7/9 85 7,12 S42 125 STAINING:sso Lines o (0.11, Therapy 0.96) P279. NUCLEI 3 Prior 8.31 1.6 5/5 65 74 POSITIVE: 60-70 Lines o (0.95, He Therapy 72.50) P27 SIGNAL S Prior O.22 5.6 88 3,8 778 16 NTENSITY:2+ Lines o (0.04, He Therapy 1.13) P65% NUCLEAR S. Prior O.13 3.3 95 2.f4 1058 16 STAINING: s5% Lines o (0.01, He-H Therapy 1.34) CD68 OVERALL No High O.23 814 411 5.19 204 POSITIVE: O-25 Timor (0.07, He Burden 0.80) CD68 OVERALL No High 2.33 10.25 12.18 349 204 POSITIVE: 51-75 Tumor (1.00, He Burden 5.41) CD68 POSITIVE No High 0.40 14, 22 432 12/15 1058 182 FOLLICULAR: Tumor (0.18, O-2S Burden 0.91) CD68 POSITIVE No High O.23 8:13 361 Sf 16 1058 182 PERIFOLLICULAR: Tumor (0.07, Ho O-2S Burden 0.78) P279. NUCLEI High O.49 1722 28O 13.25 483 248 POSITIVE: 30-SO Timor (0.24, He Burden 1.01) P27 SIGNAL High O.45 1618 122 11,17 818 248 NTENSITY: s1- Tumor (0.20, Burden 1.00)

CD68 OVERALL intermediate 0.27 7 11 343 5.15 1058 159 POSITIVE: O-25 FLIPI Score (0.08, He 0.98) CD68 POSITIVE intermediate 0.27 7, 13 361 5/17 1058 139 PERIFOLLICULAR FLIPI Score (0.08, He O-2S 0.95)

CD68 OVERALL Low FLIPI O.32 8.10 281 5.11 907 102 POSITIVE: O-25 Score (0.10, 1.01) CD68 OVERALL No Prior O.38 12.19 393 7.21 942 241 POSITIVE: 0-25 Rituximab (0.15, He Therapy 0.99)

CD68 POSITIVE No Prior O41 23,36 344 10,2S 942 210 FOLLICULAR: Rituximab (0.19, O-2S Therapy 0.87)

CD68 POSITIVE No Prior O.33 12.18 393 5.15 208 PERIFOLLICULAR: Rituximab (0.12, O-2S Therapy 0.97)

2OS 90 Prior O.40 1721 190 612 907 212 NUCLEAR Rituximab (0.16, STAINING 60-70 Therapy 1.02) US 9,322,066 B2 129 130 APPENDIX 2, TABLE 2.13-continued Treatment Free Interval by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of a 10%

Marker: Sub group HR (95% R R Vc - R Vc - R. N. Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total

CD68 OVERALL Prior O.42 18.25 115 12.20 671 2O1 POSITIVE: O-25 Rituximab (0.20, Ho Therapy 0.90)

P27 SIGNAL Prior 0.67 60,88. 272 48.85 554 210 INTENSITY: 2+ Rituximab (0.46, ke Therapy 0.98)

P65% NUCLEAR Prior O.31 1621 93 13,27 818 215 STAINING: s5% Rituximab (0.14, He Therapy 0.67) CD68 OVERALL >1 year since 0.27 17:26 269 9.28, 1058 269 POSITIVE: O-25 ast anti- (0.12, Ho ymphoma 0.63) treatinent CD68 POSITIVE >1 year since O.SO 22.34 409 16.33 907 235 FOLLICULAR: ast anti- (0.26, He O-2S ymphoma 0.97) treatinent CD68 POSITIVE >1 year since O.42 1S,24 393 11.27 1058 234 PERIFOLLICULAR: last anti- (0.19, Ho O-2S ymphoma 0.92) treatinent 20S INTENSITY European 0.59 34.52 393 25.52 907 214 CYTOPLASMIC Union (0.35, He SIGNAL: e3+ 0.99) CD68 OVERALL European O41 14, 22 234 11.25 942 211 POSITIVE: O-25 Union (0.18, He 0.92) CD68 POSITIVE European 0.37 13, 17 533 7.14 1023 185 PERIFOLLICULAR: Union (0.14, Ho >75 1.00) CD68 OVERALL sé5 years O.39 24,34 234 15,32 844 329 POSITIVE: O-25 O (0.20, s 0.76) CD68 POSITIVE sé5 years O46 33,46 190 24,41 603 292 FOLLICULAR: O (0.27, He O-2S 0.78) CD68 POSITIVE sé5 years O.SO 2O31 143 13,29 907 289 PERIFOLLICULAR: C (0.24, Ho O-2S 1.01) CD68 POSITIVE sé5 years O46 19,23 297 13,23 587 289 PERIFOLLICULAR: C (0.22, He >75 0.94) P279. NUCLE sé5 years O.S3 26,34 187 19.32 671 344 POSITIVE: O-20 C (0.29, He 0.97) P65% NUCLEAR sé5 years O.S4 Ho- 19,29 234 22.43 818 349 STAINING: siso o (0.29, 1.00) P65% POSITIVE sé5 years 0.73 103,154 343 84f142 SS7 349 CYTOPLASMIC O (0.55, ke SIGNAL: -90% 0.97) P65 INTENSITY só5 years O.74 101,152 343 82,139 SS4 349 CYTOPLASMIC O (0.55, ke SIGNAL:2- 0.99) US 9,322,066 B2 131 132 APPENDIX 2, TABLE 2.13-continued Treatment Free Interval by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub group HR (95% R R Vc - R Vc - R. N. Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total 2OS 90 Female 0.55 3S,61 427 19:49 259 NUCLEAR (0.32, He STAINING O-20 0.97) CD68 OVERALL Female O.28 1626 293 Sf 18 942 242 POSITIVE: O-25 (0.10, He 0.76) CD68 POSITIVE Female O46 23.37 307 10.22 942 217 FOLLICULAR: (0.22, He O-2S 0.96) CD68 POSITIVE Female O.28 19.28 234 5.17 942 215 PERIFOLLICULAR: (0.10, He O-2S 0.77) 2OS 90 Male O.43 14f16 110 1220 390 204 NUCLEAR (0.20, He STAINING 60-70 0.95)

CD68 OVERALL Male O.25 8.9 272 4f11 2OO POSITIVE: >75 (0.07, He 0.87) CD68 POSITIVE Asian O.13 5.6 148 2.6 30 FOLLICULAR: (0.02, He O-2S 1.17) CD68 POSITIVE Other 0.14 44 217 4.5 495 22 PERIFOLLICULAR: (0.01, Ho-H 26-SO 1.31) CD68 OVERALL White O.38 24, 35 269 16.35 907 385 POSITIVE: O-25 (0.20, H 0.73) CD68 POSITIVE White O.S9 32.48 307 25.42 671 335 FOLLICULAR: (0.35, Ho O-2S 1.00) CD68 POSITIVE White O.43 9,27 361 19:39 924 332 PERIFOLLICULAR: (0.23, H O-2S 0.83) P65% POSITIVE White O.76 113,179 363 95,162 S87 4O6 CYTOPLASMIC (0.58, |- SIGNAL: -90% 1.00)

CD68 OVERALL Ann Arbor O.36 2.16 13S 9.12 745 144 POSITIVE: O-25 Stage III (0.14, H 0.92) CD68 POSITIVE Ann Arbor 0.44 9:24, 190 13,18 603 135 FOLLICULAR: Stage III (0.21, H O-2S 0.90) P279% NUCLEI Ann Arbor O.42 4f16 275 12.15 449 149 POSITIVE: O-20 Stage III (0.17, Ho 1.00) P65 INTENSITY Ann Arbor O.38 Of 11 119 7 11. 671 151 CYTOPLASMIC Stage III (0.14, He SIGNAL: s1- 1.02)

CD68 OVERALL Ann Arbor O.39 5, 22 293 8,23 883 222 POSITIVE: O-25 Stage IV (0.16, H 0.97) CD68 POSITIVE Ann Arbor 0.37 1.15 234 8.22 924 183 PERIFOLLICULAR: Stage IV (0.14, H US 9,322,066 B2 133 134 APPENDIX 2, TABLE 2.13-continued Treatment Free Interval by Protein Expression and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of a 10%

Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total

---- O.O1 0.1 1 10

APPENDIX 2, TABLE 2.14 Treatment Free Interval by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10% Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total PSMB1 A171S: No O.80 173,276 406 1S2,266 581 S42 GiG Subgroup (0.65, 1.00) PSMB1 I2O8N: No O.80 173,276 406 1S2,266 581 S42 TT Subgroup (0.65, 1.00) PSMB1 P11A: No O.69 k 78.127 409 63,115 778 S42 C.G. Subgroup (0.50, 0.97) PSMB1 P193L: No O.80 73,276 406 S2,266 58 S42 CFC Subgroup (0.65, .00) PSMB2E49X: No O.80 73,276 406 S2,266 58 S42 GiG Subgroup (0.65, .00) PSMB2G187V: No O.80 73,276 406 S2,266 58 S42 GiG Subgroup (0.65, .00) PSMB2FL159F: No O.80 73,276 406 S2,266 58 S42 CFC Subgroup (0.65, .00) PSMBSL2O6M: No O.80 73,276 406 S2,266 58 S42 CFC Subgroup (0.65, .00) PSMB6A234D: No O.80 73,276 406 S2,266 58 S42 CFC Subgroup (0.65, .00) PSMB8.G8R: No 0.79 65,264 396 44,256 S8 S42 GiG Subgroup (0.63, 0.99) PSMB8. R141C: No O.80 73,276 406 S2,266 58 S42 CFC Subgroup (0.65, 1.00) PSMB8, V182M. No O.80 73,276 406 S2,266 58 S42 GiG Subgroup (0.65, 1.00) PSMB1 P11A: 1 Prior O.S8 32.51 480 27/55 883 231 C.G. Line of (0.35, Ke Therapy 0.97) PSMB1 P11A: 1 Prior 0.37 11.11 3O2 7.14 522 231 GiG Line of (0.14, He Therapy 0.98) US 9,322,066 B2 135 136 APPENDIX 2, TABLE 2.14-continued Treatment Free Interval by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total PSMB8.G8R: Prior O.70 69,109 S33 55,111 690 23 GiG Line of (0.49, ke Therapy 0.99) PSMB1 A171S: S Prior O.29 7/9 95 612 778 2 GiG Lines o (0.08, Ho Therapy .01) PSMB1 I2O8N: S Prior O.29 7/9 95 612 778 2 TT Lines o (0.08, He Therapy .01)

PSMB1 P193Li: S Prior O.29 7/9 95 612 778 2 CFC Lines o (0.08, Therapy .01)

PSMB2E49X: S Prior O.29 7/9 95 612 778 2 GiG Lines o (0.08, He Therapy .01)

PSMB2G187V: S Prior O.29 7/9 95 612 778 2 GiG Lines o (0.08, He Therapy .01)

PSMB2FL159F: S Prior O.29 7/9 95 612 778 2 CFC Lines o (0.08, Therapy .01) PSMBSL2O6M: 5 Prior O.29 7 9 95 6, 12 778 2 CFC Lines o (0.08, Ho Therapy .01)

PSMBSR24C: S Prior O.17 6.7 95 4.9 1058 2 CFC Lines o (0.03, Ho Therapy 0.86) PSMB6A234D: S Prior O.29 7/9 95 612 778 2 CFC Lines o (0.08, Therapy .01)

PSMB6P107A: S Prior O.29 7/9 95 612 778 2 CFC Lines o (0.08, He Therapy .01)

PSMB8.G8R: S Prior O.29 7/9 95 611 778 2 GiG Lines o (0.08, Therapy .01)

PSMB8. R141C: S Prior O.29 7/9 95 612 778 2 CFC Lines o (0.08, He Therapy .01)

PSMB8, V182M. S. Prior O.29 7/9 95 612 778 2 GiG Lines o (0.08, He Therapy .01)

PSMB9, G9E: S Prior O.29 7/9 95 612 778 2 GiG Lines o (0.08, Therapy .01)

PSMB9, V32I: S Prior O.25 7/8 88 5.11 1058 2 CFC Lines o (0.07, Therapy 0.87) PSMB1 A171S: High 0.75 108,147 234 98.149 366 296 GiG Tumor (0.57, ke Burden 0.99) PSMB1 I2O8N: High 0.75 108,147 234 98.149 366 296 TT Tumor (0.57, ke Burden 0.99) US 9,322,066 B2 137 138 APPENDIX 2, TABLE 2.14-continued Treatment Free Interval by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of a 10%

Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total

PSMB1 P11A: High O.65 49.72 218 38.6S 499 296 C.G. Tumor (0.42, e Burden 1.00) PSMB1 P193Li: High 0.75 O8,147 234 98.149 366 296 CFC Tumor (0.57, ke Burden 0.99) PSMB2E49X: High 0.75 O8,147 234 98.149 366 296 GiG Tumor (0.57, ke Burden 0.99) PSMB2/G187V: High 0.75 O8,147 234 98.149 366 296 GiG Tumor (0.57, ke Burden 0.99) PSMB2FL159F: High 0.75 O8,147 234 98.149 366 296 CFC Tumor (0.57, ke Burden 0.99) PSMB5/L206M: High 0.75 O8,147 234 98.149 366 296 CFC Tumor (0.57, ke Burden 0.99) PSMBSR24C: High O.48 19,21 148 1824 391 296 C.T Tumor (0.25, He Burden 0.93) PSMB6/A234D: High 0.75 O8,147 234 98.149 366 296 CFC Tumor (0.57, ke Burden 0.99) PSMB6P107A: High O.76 04f140 234 97.146 365 296 CFC Tumor (0.57, I Burden 1.00) PSMB8.G8R: High O.71 04f140 218 94f145 390 296 GiG Tumor (0.54, k Burden 0.94) PSMB8. R141C: High 0.75 O8,147 234 98.149 366 296 CFC Tumor (0.57, ke Burden 0.99) PSMB8/V182M: High 0.75 O8,147 234 98.149 366 296 GiG Tumor (0.57, ke Burden 0.99) PSMB9, G9E: High 0.75 O7,147 234 98.148 365 296 GiG Tumor (0.57, ke Burden 0.99) PSMB9, V32I: High O.74 O6,142 234 94f142 366 296 CFC Tumor (0.56, I Burden 0.98) PSMB1/A171S: >1 year since O.71 98.16S 547 80,166 800 331 GiG ast anti- (0.53, I ymphoma 0.95) treatinent

PSMB1 I2O8N: >1 year since O.71 98.16S 547 80,166 800 331 TT ast anti- (0.53, k ymphoma 0.95) treatinent

PSMB1 P193Li: >1 year since O.71 98.16S 547 80,166 800 331 CFC ast anti- (0.53, k ymphoma 0.95) treatinent US 9,322,066 B2 139 140 APPENDIX 2, TABLE 2.14-continued Treatment Free Interval by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total PSMB2, E49X: >1 year since O.71 98.16S S47 80,166 800 331 GiG ast anti- (0.53, I ymphoma 0.95) treatinent PSMB2/G187V: >1 year since O.71 98.16S S47 80,166 800 331 GiG ast anti- (0.53, I ymphoma 0.95) treatinent PSMB2/L159F: >1 year since O.71 98.16S S47 80,166 800 331 CFC ast anti- (0.53, I ymphoma 0.95) treatinent PSMB5/L206M: >1 year since O.71 98.16S S47 80,166 800 331 CFC ast anti- (0.53, I ymphoma 0.95) treatinent PSMBSR24C: >1 year since 0.72 81,137 507 68. 141 843 331 CFC ast anti- (0.52, k ymphoma 1.00) treatinent

PSMB6/A234D: >1 year since O.71 98.16S S47 80,166 800 331 CFC ast anti- (0.53, I ymphoma 0.95) treatment

PSMB6/P107A: >1 year since 0.72 95,159 561 79,162 800 331 CFC ast anti- (0.53, I ymphoma 0.97) treatinent PSMB8, G8R: >1 year since O.71 91.155 562 74,159 843 331 GiG ast anti- (0.52, I ymphoma 0.96) treatinent PSMB8/R141C: >1 year since O.71 98.16S S47 80,166 800 331 CFC ast anti- (0.53, I ymphoma 0.95) treatinent

PSMB8/V182M: >1 year since O.71 98.16S S47 80,166 800 331 GiG ast anti- (0.53, | ymphoma 0.95) treatinent

PSMB9, G9E: >1 year since O.71 97,163 547 80.16S 800 331 GiG ast anti- (0.53, I ymphoma 0.95) treatinent PSMB9, V32I: >1 year since O.70 96.16O SO7 76.158 800 331 CFC ast anti- (0.51, ke ymphoma 0.94) treatinent

PSMB1AP11A: European O.S6 34.f49 309 29,51 844 241 C.G. Union (0.34, He 0.93) PSMB1/A171S: sé5 years O.74 130,198 344 112,194 SS7 392 GiG old (0.57, ke 0.95) PSMB1/I208N: sé5 years O.74 130,198 344 112,194 SS7 392 TT old (0.57, I 0.95)

US 9,322,066 B2 143 144 APPENDIX 2, TABLE 2.14-continued Treatment Free Interval by Germline Genetic Variant and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of a 10%

Marker: Sub group HR (95% R R Vc - R Vc - R. N Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median Total

PSMB1AP11A: White O.69 67.111 432 54,97 818 473 C.G. (0.48, s 0.99) PSMB1AP11A: White 0.57 27.32 272 25,39 449 473 GiG (0.33, e 0.98)

PSMB1AP11A: Ann Arbor O.S3 29:44 2O6 16,28 818 172 C.G. Stage III (0.28, He 0.98)

O.O1 0.1 1 10 100

APPENDIX 2, TABLE 2.15 Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of e10% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal BCL2A43G: No 0.67 104,161 344 84,154 649 315 MIND Subgroup (0.50, He 0.90) BCL2.C.-11C > T: No 0.72 127,192 361 102,179 587 390 MIND Subgroup (0.56, He 0.94) BCL2.E.29K: No O.65 98.1SO 343 77,139 649 318 MIND Subgroup (0.48, He 0.87) BCL2 P46L: No 0.72 86,135 362 77,139 595 314 MIND Subgroup (0.53, He 0.98) BCL2.P59S: No 0.73 92,145 362 84f147 629 309 MIND Subgroup (0.54, He 0.99) BCL2/Q52P: No 0.73 97,153 361 88,154 629 307 MIND Subgroup (0.55, He 0.97) BCL2iR106H: No O.68 08.162 344 84f151 629 372 MIND Subgroup (0.51, He 0.90) NOTCHAG A1702P: No O.74 13,175 361 93.16.1 589 337 MIND Subgroup (0.57, He 0.98) NOTCHAI1681N: No 0.75 13,175 361 94.162 587 337 MIND Subgroup (0.57, He 0.99) NOTCHAL1679P: No 0.75 13,175 361 94.162 587 337 MIND Subgroup (0.57, He 0.99) NOTCH/L1679Q: No 0.75 13,175 361 94.162 587 337 MIND Subgroup (0.57, He 0.99) US 9,322,066 B2 145 146 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal NOTCHP2513Li: No O.71 90,140 360 67.123 649 361 MIND Subgroup (0.52, He 0.97) NOTCHAP2S1SFS4: No O.71 114,176 361 92,166 629 344 MIND Subgroup (0.54, Ho 0.94) NOTCH/Q246OX: No O.74 98.151 344 82,143 589 324 MIND Subgroup (0.55, Ho 0.99) NOTCHSX26DEL: No O.76 112,175 363 102,178 595 361 MIND Subgroup (0.58, He 0.99) NOTCHSX26INS: No 0.75 114,179 363 103,180 595 361 MIND Subgroup (0.58, He 0.99) NOTCHSX28DEL: No O.74 128,194 361 109,189 583 400 MIND Subgroup (0.57, He 0.96) NOTCHSX28INS: No 0.72 132,198. 361 112,195 583 4O1 MIND Subgroup (0.56, He 0.93) BCL2A43G: Prior 0.55 46.72 409 30.67 907 39 MIND Line o (0.35, He Therapy 0.88) BCL2.C.-11C > T: Prior O.63 57,86 427 38.78 827 70 MIND Line o (0.42, He Therapy 0.95) BCL2.E.29K: Prior O.S6 41.64 406 29.62 828 39 MIND Line o (0.34, He Therapy 0.90) BCL2 P46L: Prior O.61 37.58 449 30.63 828 40 MIND Line o (0.38, He Therapy 1.00) BCL2 P46S: Prior O.62 40.64 409 33.68 828 38 MIND Line o (0.39, He Therapy 0.99) BCL2/Q52P: Prior O.63 42.67 409 35.71 827 38 MIND Line o (0.40, He Therapy 0.99) BCL2iR106H: Prior O.S9 49,73 409 33.68 828 62 MIND Line o (0.38, He Therapy 0.92) NOTCHP2515FS4: Prior O.64 48,75 432 35/72 827 49 MIND Line o (0.41, Therapy 0.99) NOTCHSX28INS: Prior O.65 56.84 427 45.88 673 78 MIND Line o (0.44, Ho Therapy 0.96) BCL2A43G: No High O.61 42.78 632 27.64 42 MIND Tumor (0.38, He Burden 1.00) BCL2iR106H: No High O.61 46.79 549 26.58 1023 63 MIND Tumor (0.38, Ho Burden 0.99) NOTCHP2513Li: No High O.45 17.26 452 9,21 63 MD Tumor (0.20, He Burden 1.01) US 9,322,066 B2 147 148 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

NOTCHSX28INS: No High O.64 55.98 592 37.82 924 184 MIND Tumor (0.42, Ho Burden 0.98) BCL2A43G: High O.62 62.83 234 57,90 379 173 MIND Tumor (0.43, He Burden 0.89) BCL2.C.-11C > T: High O.68 76.101 234 69,107 326 217 MIND Tumor (0.49, Burden 0.94) BCL2.E.29K: High 0.57 60,79 234 S2,82 392 74 MIND Tumor (0.40, Burden 0.84) BCL2 P46L: High O.S9 51,67 234 51.83 379 73 MIND Tumor (0.40, He Burden 0.87) BCL2 P46S: High O.64 52.73 234 S6,89 337 70 MIND Tumor (0.44, He Burden 0.94) BCL2.P59L: High 0.67 52/72 254 SO.80 3.18 72 MIND Tumor (0.45, He Burden 0.99) BCL2P59S: High O.63 53,73 254 54.87 379 70 MIND Tumor (0.43, He Burden 0.93) BCL2/Q52P: High O.63 55/76 234 58,92 379 68 MIND Tumor (0.44, He Burden 0.92) BCL2iR106H: High O.61 62.83 234 58.93 392 209 MIND Tumor (0.43, He Burden 0.88) NOTCH/G A1702P: High O.64 67.90 234 62.96 379 187 MIND Tumor (0.45, He Burden 0.91) NOTCHAI1681N: High O.65 67.90 234 63.97 379 187 MIND Tumor (0.46, He Burden 0.92) NOTCH/L1586Q: High 0.67 56,78 254 6094 379 172 MIND Tumor (0.47, He Burden 0.97) NOTCHAL1597H: High 0.67 56,78 254 6094 379 72 MIND Tumor (0.47, Burden 0.97) NOTCHAL1679P: High O.65 67.90 234 63.97 379 87 MIND Tumor (0.46, He Burden 0.92) NOTCH/L1679Q: High O.65 67.90 234 63.97 379 87 MIND Tumor (0.46, Burden 0.92) NOTCHAL2458V: High O.66 61.83 234 61,95 379 78 MIND Tumor (0.46, Burden 0.94) NOTCHP2513Li: High O.S6 58,76 234 43.71 392 98 MIND Tumor (0.37, He Burden 0.83) US 9,322,066 B2 149 150 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal NOTCH/P2515FS4: High O.63 67.90 234 61.98 379 89 MIND Tumor (0.45, He Burden 0.90) NOTCH/Q2441X: High O.66 61.83 234 61,95 379 78 MIND Tumor (0.46, Ho Burden 0.94) NOTCH/Q246OX: High O.63 59.79 234 54.87 392 81 MIND Tumor (0.43, He Burden 0.91) NOTCHAR1599P: High 0.67 56,78 254 6094 379 72 MIND Tumor (0.47, He Burden 0.97) NOTCHV1579E: High 0.67 56,78 254 6094 379 72 MIND Tumor (0.47, He Burden 0.97) NOTCH/X26DEL: High O.69 63,88 254 68,107 379 98 MIND Tumor (0.49, He Burden O.98 NOTCHSX26INS: High O.68 64.89 254 68,107 379 98 MIND Tumor (0.49, He Burden 0.96) NOTCH/X28DEL: High 0.67 74,99 234 72,111 326 216 MIND Tumor (0.48, He Burden 0.92) NOTCHSX28INS: High 0.67 77,100 190 75,113 3.18 217 MIND Tumor (0.49, He Burden 0.92) BCL2A43G: intermediate O.60 34,53 362 26,52 778 05 MIND FLIPI (0.36, He Score 1.00) BCL2.C.-11C > T: intermediate O.60 44f66 362 33.63 778 36 MIND FLIPI (0.38, He Score 0.95) BCL2iR106H: intermediate O.S2 40.55 361 28,54 671 30 MIND FLIPI (0.32, Ho Score 0.84) BCL2.E.29K: No Prior O.66 57,93 393 41,76 718 86 MIND Rituximab (0.44, He Therapy 0.98) BCL2.P59L: MD No Prior O.17 3.5 217 7, 13 779 78 Rituximab (0.03, Ho Therapy 0.91) BCL2A43G: Prior O.62 44f63 287 37,69 589 32 MIND Rituximab (0.40, He Therapy 0.96) BCL2.C.-11C > T: Prior O.60 57/78 254 43,77 554 65 MIND Rituximab (0.40, He Therapy 0.89) BCL2.E.29K: Prior O.62 41.57 287 36.63 581 32 MIND Rituximab (0.39, He Therapy 0.97) BCL2.P59S: Prior O.63 39 S6 307 37,69 589 33 MIND Rituximab (0.40, Ho Therapy 0.98) BCL2iR106H: Prior 0.57 50.67 272 39.72 581 59 MIND Rituximab (0.38, He Therapy 0.88) US 9,322,066 B2 151 152 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

NOTCHAG A1702P: Prior O.62 51/72 288 41,74 589 47 MIND Rituximab (0.41, He Therapy 0.94) NOTCHAI1681N: Prior O.64 51/72 288 42.75 581 47 MIND Rituximab (0.42, He Therapy 0.96) NOTCHAL1679P: Prior O.64 51/72 288 42.75 581 47 MIND Rituximab (0.42, He Therapy 0.96) NOTCH/L1679Q: Prior O.64 51/72 288 42.75 581 47 MIND Rituximab (0.42, He Therapy 0.96) NOTCHAL2458V: Prior O.63 47.66 3O2 42.75 581 41 MIND Rituximab (0.41, He Therapy 0.96) NOTCHP2513Li: Prior O.S9 39.57 307 25,51 649 S4 MIND Rituximab (0.36, He Therapy 0.98) NOTCHAP2S1SFS4: Prior O.63 51/72 288 42.77 581 49 MIND Rituximab (0.42, He Therapy 0.96) NOTCH/Q2441X: Prior O.63 47.66 3O2 42.73 581 41 MIND Rituximab (0.42, He Therapy 0.97) NOTCH/Q246OX: Prior O.60 45,62 288 36.66 589 43 MIND Rituximab (0.39, He Therapy 0.93) NOTCHSX26INS: Prior O.66 52.74 288 47.84 581 58 MIND Rituximab (0.45, He Therapy 0.98) NOTCHSX28DEL: Prior O.64 60,83 28O 4986 554 75 MIND Rituximab (0.43, He Therapy 0.93) NOTCHSX28INS: Prior O.62 62.86 28O SO.88 554 75 MIND Rituximab (0.43, He Therapy 0.91) BCL2A43G: >1 year since O.65 57,91 449 44.94 716 85 MIND ast anti- (0.44, He ymphoma 0.96) treatinent BCL2/C-11C > T: >1 year since O.64 73,112 419 55, 112 716 233 MIND ast anti- (0.45, Ho ymphoma 0.90) treatinent

BCL2, E29K: >1 year since O.60 52.82 409 39.83 718 185 MIND ast anti- (0.40, He ymphoma 0.91) treatinent

BCL2iR106H: >1 year since O.S8 65.96 392 45.94 716 217 MIND ast anti- (0.40, He ymphoma 0.85) treatinent NOTCHP2513Li: >1 year since O.61 50,78 419 37/81 716 213 MIND ast anti- (0.40, He ymphoma 0.94) treatinent US 9,322,066 B2 153 154 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal NOTCH/P2515FS4: >1 year since 0.66 64,102 452 48,102 716 205 MIND ast anti- (0.45, He ymphoma 0.96) treatinent NOTCH/X28INS: >1 year since 0.66 73.113 432 59,119 709 237 MIND ast anti- (0.47, He ymphoma 0.93) treatinent BCL2A43G: European O.S4 41,66 SO7 32.72 942 38 MIND Union (0.34, He 0.86) BCL2C.-11C > T: European O.S8 53,79 363 41.83 843 69 MIND Union (0.38, He 0.87) BCL2.E.29K: European O.60 37,61 393 32.67 907 38 MIND Union (0.37, 0.96) BCL2 P46L: European O.S9 34;S6 393 29,64 1023 39 MIND Union (0.36, He 0.97) BCL2 P46S: European O.60 37,61 362. 33.68 907 36 MIND Union (0.37, He 0.96) BCL2.P59S: European O.61 36.60 363 34f69 843 36 MIND Union (0.38, He 0.97) BCL2/Q52P: European O.61 38.63 362 36,73 843 36 MIND Union (0.38, He 0.96) BCL2iR106H: European O.S8 44f6S 393 35,7O 844 60 MIND Union (0.37, He 0.90) NOTCHG A1702P: European O.62 44f72 393 36,74 843 47 MIND Union (0.40, He 0.96) NOTCHI1681N: European O.64 44f72 393 37,75 843 47 MIND Union (0.41, He 0.99) NOTCHL1679P: European O.64 44f72 393 37,75 843 47 MIND Union (0.41, He 0.99) NOTCH/L1679Q: European O.64 44f72 393 37,75 843 47 MIND Union (0.41, He 0.99) NOTCHP2513L: European O.48 39,59 362 22.52 942 59 MIND Union (0.28, He 0.81) NOTCHAP2S1SFS4: European O.S9 48.77 393 36.77 907 55 MIND Union (0.38, He 0.91) NOTCHX26INS: European O.64 46, 74 411 4284 844 59 MIND Union (0.42, He 0.98) NOTCHX28DEL: European O.63 53.82 393 45.87 843 77 MIND Union (0.42, He

US 9,322,066 B2 157 158 APPENDIX 2, TABLE 2.15-continued

Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. (All Reported Groups are Significant (ps 0.05) and at a Frequency of >10%)

Marker: Sub HR (95% Vc - R Vc - R. group Mar ker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

Female O.62 51.86 393 31,68 1023 MN D (0.40, 0.98)

Male O.S6 45.59 215 587 35 MN D (0.37, 0.84)

Male O.63 57.77 234 58.93 534 82 D (0.44, 0.92)

Male O.S2 43/55 190 43.68 587 38 D (0.34, 0.79)

Male O.11 44 244 6.10 716 34 (0.02, 0.63)

Male O.64 39.53 254 50,78 543 31 MN D (0.42, 0.98)

BCL2. Male O.62 SOf 66 254 49.81 543 69 MN (0.42, 0.93)

NO TC fo A1702P: Male O.61 47.63 90 S1,81 581 45 MN D (0.41, 0.91)

NO TC AI1681N: Male O.62 47.63 90 S2,82 543 45 MN D (0.42, 0.92)

NO Male O.62 47.63 90 S2,82 543 45 MN (0.42, 0.92)

NO Male O.62 44,59 90 S1,80 543 39 MN (0.41, 0.93)

NO P2513Li: Male O.S4 41.54 90 37,63 629 55 MN (0.35, 0.85)

NO Male O.S8 47.62 234 S1,84 587 48 MN (0.39, 0.87)

Male O.62 44,59 190 51.79 543 39 MN (0.41, 0.93)

NO TC Male O.14 3.3 55 8,12 SO4 42 MD (0.03, 0.71) US 9,322,066 B2 159 160 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. (All Reported Groups are Significant (ps 0.05) and at a Frequency of >10%)

Marker: Sub HR (95% R Vc - R Vc - R. group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

NOTC Male O.62 42.57 190 43,70 581 42 MIND (0.40, 0.95)

NOTCHSX26DEL: Male O.S9 254 58.95 583 62 MIND (0.40, 0.87)

NOTCHSX26INS: Male 0.57 SOf 66 234 58.95 583 62 MIND (0.39, 0.84)

NOTCHSX28DEL: Male O.65 56,75 254 61.97 543 8O MIND (0.45, 0.94)

NOTCHSX28INS: Male O.64 59.78 234 63,100 534 81 MIND (0.45, 0.91)

BCL2A43G: O.68 89. 40 361 76.139 671 279 MIND (0.50, 0.93)

BCL2.C.-11C > T: 0.72 110. 68 362 90,158 589 342 MIND (0.54, 0.95)

BCL2.E.29K: O.64 83, 29 344 70,127 671 282 MIND (0.47, 0.88)

BCL2 P46L: O.70 74, 17 363 68,125 649 279 MIND (0.50, He 0.97)

BCL2 P46S: O.71 81, 30 363 74,131 671 275 MIND (0.52, He 0.98)

BCL2.P59S: O.70 81, 28 363 73.130 671 274 MIND (0.51, He 0.96) BCL2/Q52P: O.70 85. 35 362 77,137 671 272 MIND (0.51, He 0.96)

BCL2iR106H: O.70 93. 40 361 77,135 629 327 MIND (0.51, 0.94)

NOTCHAG A1702P: White 0.73 97. 52 362 82,143 629 295 MIND (0.55, 0.99)

NOTCHAI1681N: White 0.73 97. 52 362 82,143 629 295 MIND (0.55, 0.99)

NOTCHAL1679P: White 0.73 97. 52 362 82,143 629 295 MIND (0.55, 0.99) NOTCH/L1679Q: White 0.73 97. 52 362 82,143 629 295 MIND (0.55, 0.99) US 9,322,066 B2 161 162 APPENDIX 2, TABLE 2.15-continued

Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. (All Reported Groups are Significant (ps 0.05) and at a Frequency of >10%)

Marker: Sub HR (95% R R Vc - R Vc - R. group Mar ker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

NO TC P2513Li: White O.69 78.122 362 59. 673 319 MN (0.49, 0.96)

NO TC White O.69 99.154 363 81. 48 671 MN (0.51, 0.93)

NO TC White O.71 85,131 361 73. 27 589 284 MN (0.52, 0.98)

NO TCHSX26DEL: White 0.75 99.155 363 91, 58 649 321 MN (0.56, 0.99)

NO TCHSX26INS: White O.74 101,159 392 92. 60 649 321 MN (0.56, 0.98)

NO TCHSX28DEL: White 0.75 111,169 362 98. 67 587 351 MN (0.57, 0.98)

NO TCHSX28INS: White 0.72 115,174 362 99. 71 589 352 (0.55, 0.94)

Ann Arbor O.62 48,71 309 42,83 495 S4 Stage IV (0.41, 0.93)

Ann Arbor O.66 6084 309 S4,100 471 93 Stage IV (0.46, 0.95)

Ann Arbor O.60 46.67 307 490 S4 MN Stage IV (0.39, 0.93)

Ann Arbor O.19 3.5 145 8,12 49 Stage IV (0.03, 1.00)

Ann Arbor O.60 50, 72 307 42,83 490 86 MN Stage IV (0.40, 0.91)

NO CCHAG A1702P: Ann Arbor O.62 52.75 287 47/87 487 63 MN D Stage IV (0.42, 0.93)

NO TCHAI1681N: Ann Arbor O.64 52.75 287 48.88 487 63 MN D Stage IV (0.43, 0.94)

NO TCHAL1679P: Ann Arbor O.64 52.75 287 48.88 487 63 MN D Stage IV (0.43, 0.94)

NO Ann Arbor O.64 52.75 287 48.88 487 63 MN Stage IV (0.43, 0.94) US 9,322,066 B2 163 164 APPENDIX 2, TABLE 2.15-continued Treatment-Free Interval by Somatic Mutation and by Covariate, IRC Review. All Reported Groups are Significant (ps 0.05) and at a Frequency of 210% Marker: Sub HR (95% R R Vc - R Vc - R group Marker: Level Subgroup CI) HR (log scale) Evti N Median Evti N Median NTotal

NOTCHAL2458V: Ann Arbor O.65 46/68 293 46.86 490 S4 MIND Stage IV (0.43, Ho 0.98) NOTCHP2513Li: Ann Arbor 0.57 42.61 307 34,69 629 76 MIND Stage IV (0.36, He 0.90) NOTCHAP2S1SFS4: Ann Arbor O.60 51.74 293 46.91 495 67 MIND Stage IV (0.40, 0.90) NOTCH/Q2441X: Ann Arbor O.65 46/68 293 46,85 490 S4 MIND Stage IV (0.43, c 0.98) NOTCH/Q246OX: Ann Arbor O.63 43.64 307 41,79 495 56 MIND Stage IV (0.41, 0.96) NOTCHX26DEL: Ann Arbor O.65 53,76 317 S4,101 490 8O MIND Stage IV (0.45, He 0.95) NOTCHSX26INS: Ann Arbor O.65 54f79 309 S4,101 490 8O MIND Stage IV (0.44, 0.94) NOTCHX28DEL: Ann Arbor O.66 61,85 307 S6,103 483 95 MIND Stage IV (0.46, He 0.94) NOTCHSX28INS: Ann Arbor O.62 64;87 293 57.105 183 96 MIND Stage IV (0.43, o 0.89)

------0.01 0.02 0.05 01 0.2 0.5 1

APPENDIX 3

Pair-Wise Combinations of Markers 45

The following table outlines the data for all significant pair wise combinations. Note: Selected=Biomarker positive, Not 50 Selected=Biomarker negative

65 US 9,322,066 B2 165 166

US 9,322,066 B2 167 168

US 9,322,066 B2 171 172

US 9,322,066 B2 177 178