IDENTIFICATION OF PATIENTS IN NEED OF PD-L1 INHIBITOR COTHERAPY

Abstract

The present invention relates to means and methods for determining whether a patient is in need of a PD-L1 inhibitor cotherapy. A patient is determined to be in need of the PD-L1 inhibitor cotherapy if a low or absent ER expression level and an expression level of programmed death ligand 1 (PD-L1) that is increased in comparison to a control is measured in vitro in a sample from the patient. The patient is undergoing therapy comprising a modulator of the HER2/neu (ErbB2) signaling pathway (like Trastuzumab) and a chemotherapeutic agent (like dodetaxel) or such a therapy is contemplated for the patient. Also provided herein are means and methods for treating a cancer in a cancer patient for whom therapy comprising a modulator of the HER2/neu (ErbB2) signaling pathway (like Trastuzumab) and a chemotherapeutic agent (like dodetaxel) is contemplated, wherein the patient is to receive PD-L1 inhibitor cotherapy.

Claims

1. A method of determining the need of a cancer patient for a PD-L1 inhibitor cotherapy, (i) wherein therapy comprising a modulator of the HER2/neu (ErbB2) signaling pathway and a chemotherapeutic agent is contemplated for the patient or (ii) wherein the patient is undergoing therapy comprising a modulator of the HER2/neu (ErbB2) signaling pathway and a chemotherapeutic agent, the method comprising the steps of a) measuring in vitro in a sample from said patient the expression level of Estrogen receptor (ER) and of programmed death ligand 1 (PD-L1), b) determining a patient as being in need of a PD-L1 inhibitor cotherapy if a low or absent ER expression level and an expression level of programmed death ligand 1 (PD-L1) that is increased in comparison to a control is measured in step (a).

2-83. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0230] FIG. 1 provides a schematic of the HER2 protein structure, and amino acid sequences for Domains I-IV, respectively) of the extracellular domain thereof (SEQ ID NOS. 22-25, respectively, in order of appearance).

[0231] FIGS. 2A and 2B depict alignments of the amino acid sequences of the variable light (V.sub.L) (FIG. 2A) and variable heavy (V.sub.H) (FIG. 2B) domains of murine monoclonal antibody 2C4 (SEQ ID Nos. 26 and 27, respectively); V.sub.L and V.sub.H domains of variant 574/Pertuzumab (SEQ ID Nos. 28 and 29, respectively), and human V.sub.L and V.sub.H consensus frameworks (hum κ1, light kappa subgroup I; humIII, heavy subgroup III) (SEQ ID Nos. 30 and 31, respectively). Asterisks identify differences between variable domains of Pertuzumab and murine monoclonal antibody 2C4 or between variable domains of Pertuzumab and the human framework. Complementarity Determining Regions (CDRs) are in brackets.

[0232] FIGS. 3A and 3B show the amino acid sequences of Pertuzumab light chain (FIG. 3A; SEQ ID NO: 32) and heavy chain (FIG. 3B; SEQ ID NO: 33). CDRs are shown in bold. Calculated molecular mass of the light chain and heavy chain are 23,526.22 Da and 49,216.56 Da (cysteines in reduced form). The carbohydrate moiety is attached to Asn 299 of the heavy chain.

[0233] FIGS. 4A and 4B show the amino acid sequences of Trastuzumab light chain (FIG. 4A; SEQ ID NO: 34) and heavy chain (FIG. 4B; SEQ ID NO: 35), respectively. Boundaries of the variable light and variable heavy domains are indicated by arrows.

[0234] FIGS. 5A and 5B depict a variant Pertuzumab light chain sequence (FIG. 5A; SEQ ID NO: 36) and a variant Pertuzumab heavy chain sequence (FIG. 5B; SEQ ID NO: 37), respectively.

[0235] FIG. 6: FIGS. 6A and 6B show known mRNA transcripts and position of the relevant AFFYMETRIX probe set target regions for gene CD274. Exons are shown as grey bold rectangles, junction regions are indicated by thin horizontal lines. Probe sets with their sequence mapped against mRNA sequences are shown as black bold rectangles. Provided coordinates are genomic coordinates on chromosome 9.

[0236] FIG. 7: FIGS. 7A and 7B show known mRNA transcripts and position of the relevant AFFYMETRIX probe set target regions for gene IFNG. Exons are shown as grey bold rectangles, junction regions are indicated by thin horizontal lines. Probe sets with their sequence mapped against mRNA sequences are shown as black bold rectangles. Provided coordinates are genomic coordinates on chromosome 12.

[0237] FIG. 8 shows the distribution of the expression of genes IFNG and CD274 in the samples of ER− and ER-30 populations. Symbol types correspond to the final pCR status (solid: pCR achieved, open-pCR not achieved).

[0238] FIG. 9A is a box plot of expression of gene CD274 for ER− responders (pCR=YES) and nonresponders (pCR=NO). On the right the histograms of expression for both categories are provided. FIG. 9B shows a distribution of t-test statistics (HO hypothesis of no difference). The vertical mark indicates the actual value found in the involved sample. The area of the shaded regions corresponds to the alpha level.

[0239] FIG. 10A is a box plot of expression of gene IFNG for ER− responders (pCR=YES) and nonresponders (pCR=NO). On the right the histograms of expression for both categories are provided. FIG. 10B shows the distribution of t-test statistics (HO hypothesis of no difference). The vertical mark indicates the actual value found in the involved sample. The area of the shaded regions corresponds to the alpha level.

[0240] FIG. 11A is a box plot of expression of gene CD274 for ER+ responders (pCR=YES) and nonresponders (pCR=NO). On the right the histograms of expression for both categories are provided. FIG. 11B shows a distribution of t-test statistics (HO hypothesis of no difference). The vertical mark indicates the actual value found in the involved sample. The area of the shaded regions corresponds to the alpha level.

[0241] FIG. 12A is a box plot of expression of gene IFNG for ER+ responders (pCR=YES) and nonresponders (pCR=NO). On the right the histograms of expression for both categories are provided. FIG. 12B shows a distribution of t-test statistics (HO hypothesis of no difference). The vertical mark indicates the actual value found in the involved sample. The area of the shaded regions corresponds to the alpha level.

[0242] FIG. 13 is the receiver operating characteristic of the final logistic regression model for ER− population. Positive level is taken to be the positive response status (pCR=YES).

[0243] FIG. 14 is a LIFT curve of the final logistic regression model for ER− population. Positive level is taken to be the positive response status (pCR=YES). The Y-axis displays the ratio of how rich the portion of the population is in the chosen response level (upper curve corresponds to pCR=YES) compared to the rate of that response level as a whole.

[0244] FIG. 15 is an example of predicted clinical response status for ER− population. Shown is predicted profile of response as controlled by patient age, Cancer type, pN status, and expression of both genes involved. The actual predicted pCR probability (which is equal to 0.443) is given for NO LABC patient around 60 y. old and with expression in both genes around median values.

[0245] FIG. 16 is the receiver operating characteristic of the final logistic regression model for ER+ population. Positive level is taken to be the positive response status (pCR=YES).

[0246] FIG. 17 shows the distribution of age of ER− patients.

[0247] FIG. 18 shows the distribution of age of ER+ patients.

[0248] FIG. 19 shows a decision tree view on expression of IFNG and CD274 genes predicting clinical response in ER patients. The first two splits required to explain pCR are the ones wrt to IFNG and CD274.

[0249] The Example illustrates the invention.

Example 1: Cancer Patients Undergoing HER2 Targeted Therapy and Chemotherapy Benefit from PD-L1 Inhibitor Cotherapy, if the Expression Level of ER is Low or Absent (ER Negative) and if PD-L1 Expression Level is Increased

[0250] Estimation of gene expression was performed with the help of R Bioconductor package ‘affy’, R version 2.15.0. All exploratory analyses and predictive models were made using SAS JMP ver. 10.0 48 HER2+, ER+ and 39 HER2+, ER− breast cancer biopsies were obtained from NeoSphere clinical trial. The samples had been taken at diagnosis from patients afterwards treated with Docetaxel and Trastuzumab in a neo-adjuvant setting. The distribution of main clinical covariates at base line, as well as of clinical response (as assessed at the surgery) in the involved population is as follows:

ER negative samples:

Patient Age (see FIG. 17)

[0251]

TABLE-US-00016 Quantiles 100.0% maximum 72 99.5% 72 97.5% 71.55 90.0% 64 75.0% quartile 54 50.0% median 50.5 25.0% quartile 44.25 10.0% 39 2.5% 34.675 0.5% 34 0.0% minimum 34 Level Count Prob Cancer Type IBC  2 0.04167 LABC 22 0.45833 OPERABLE 24 0.50000 Total 48 1.00000 pT (pathologic staging of Tumor) T2 18 0.37500 T3 15 0.31250 T4 15 0.31250 Total 48 1.00000 pN (pathologic staging of nodes) N0 12 0.25000 N1 36 0.75000 Total 48 1.00000 G (Grade) G1  1 0.02083 G2 15 0.31250 G3 16 0.33333 NA 16 0.33333 Total 48 1.00000
ER positive samples:

Patient Age (see FIG. 18)

[0252]

TABLE-US-00017 Quantiles 100.0% maximum 74 99.5% 74 97.5% 74 90.0% 65 75.0% quartile 57 50.0% median 50 25.0% quartile 43 10.0% 40 2.5% 32 0.5% 32 0.0% minimum 32 Level Count Prob Cancer Type IBC  5 0.12821 LABC  8 0.20513 OPERABLE 26 0.66667 Total 39 1.00000 pT T2 15 0.38462 T3 16 0.41026 T4  8 0.20513 Total 39 1.00000 pN N0 11 0.28205 N1 28 0.71795 Total 39 1.00000 G G2 13 0.33333 G3 10 0.25641 NA 16 0.41026 Total 39 1.00000

[0253] Contingency Analysis of pathological complete response (pCR) By estrogen receptor status (ER)

TABLE-US-00018 Count Row % pCR = NO pCR = YES ER = ER− 27 21 48 56.25 43.75 ER = ER+ 33 6 39 84.62 15.38 60 27 87

Gene Expression Profiling

[0254] The tumor biopsy samples were profiled for gene expression on AFFYMETRIX HG-U133Plus 2 whole Human Genome microarray platform. Roche HighPure RNA extraction, NuGen amplification and standard AFFYMETRIX hybridization and scanning protocols were used. All array scans passed standard AFFYMETRIX QC.

[0255] Robust Multiarray algorithm (RMA) was used for preprocessing of raw signals (Irizarry et al, 2003. available at ncbi.nlm.nih.gov/pubmed/12925520). All probe sets available for the genes of interest were retrieved as reported below. For gene CD274, when several probe sets were available to represent this gene, the probe set with the probe set with the highest average expression value (defined as an arithmetical average of expression of a given probe set) was selected to represent the gene:

CD274 (PDLL)

[0256] 223834_at selected for PDL1
227458_at

[0257] The selected probe set corresponds to the last exon/3′UTR of the gene and captures all known RefSEq mRNAs (see FIGS. 6A and 6B)

IFNG

[0258] 210354_at

[0259] This probe set also represents the last exon/3′UTR of the gene and captures all known RefSEq mRNAs (see FIGS. 7A and 7B)

[0260] FIG. 8 shows joint distribution of the expression of the above genes in the samples of both ER- and

[0261] ER− populations. Symbol types correspond to the final pCR status (solid: pCRachieved, open-pCR not achieved).

[0262] More details on distribution of CD274 and IFNG expression across ER and pCR strata can be found in Appendix I.

[0263] For every ER subpopulation, a logistic regression model was constructed that relates expression of the selected genes with clinical response adjusted for patient age, cancer type, and nodal status: Response˜Patient.Age+Cancer.Type+pN+CD274+IFNG

1. ER− Population.

[0264] Summarized model output is given below. Odds ratios are (OR) provided per unit change of biomarker value. As the expression values are given on log 2 scale, one unit change would correspond to 2-fold overexpression. For details see Appendix.

TABLE-US-00019 ER− population LR test Term OR (95% CI) p-value CD274 5.2 (1.5; 26.7) 0.008 IFNG 0.30 (0.10; 0.74) 0.007 Patient Age 0.24 Cancer Type 0.91 pN 0.87

[0265] The final model for predicting probability for a particular patient to respond to the treatment includes expression of CD274 and IFNG and looks like:

p(pCR)=−3.737+1.607*CD274-1.069*IFNG

2. ER+ Population.

[0266] Summarized model output is given below. Odds ratios are (OR) provided per unit change of biomarker value. As the expression values are given on log 2 scale, one unit change would correspond to 2-fold overexpression. For details see Appendix.

TABLE-US-00020 ER+ population LR test Term OR (95% CI) p-value CD274 0.93 IFNG 0.23 Patient Age 0.34 Cancer Type 0.39 pN 0.92

[0267] The role of PDL1 expression is evident in ER− subpopulation of HER2+ breast cancer patients that underwent combinational treatment with Trastuzumab and chemotherapy in the neoadjuvant setting. Namely, overexpression of PDL1 at diagnosis corresponds to a lower rate of response to neoadjuvant therapy (i.e. a lower rate of response to combinational treatment with Trastuzumab and chemotherapy). This holds irrespective of patient age, cancer type, or lymph node status. A baseline assessment of gene expression of either of the two biomarkers, PDL1 and INFG, respectively, allows to identify if a patient is likely to experience a greater benefit if a PDL-1 targeted therapy is added to Trastuzumab and chemotherapy.

[0268] The following relates to a cut-off value allowing determining a patient as being in need of a PD-L1 inhibitor cotherapy in accordance with the present invention.

[0269] If a gene expression analysis gives a result for IFNG expression higher or equal to 4.8 no combination treatment (HER2-targeted and PDL1-targeted) is recommended and no further PDL1 assessment would be necessary. If a gene expression analysis gives a result for IFNG lower than 4.8 a parallel assessment of PDL-1 is necessary. If PDL-1 gene expression analysis then gives a result of higher or equal to 5.3 a combination treatment (HER2-targeted and PDL1-targeted) is recommended (see FIG. 19).

TABLE-US-00021 APPENDIX I ER− subpopulation Oneway Analysis of CD274 Expression By pCR ER = ERneg (see FIG. 9A) t Test YES-NO Assuming unequal variances Difference −0.32948 t Ratio −1.94171 Std Err Dif 0.16969 DF 45.11513 Upper CL Dif 0.01226 Prob > |t| 0.0584 Lower CL Dif −0.67122 Prob > t 0.9708 Confidence 0.95 Prob < t 0.0292* The results are also shown in FIG. 9B. Oneway Analysis of IFNG Expression By pCR ER = ERneg The results are shown in FIG. 10A. t Test YES-NO Assuming unequal variances Difference 0.58405 t Ratio 2.044225 Std Err Dif 0.28571 DF 30.21429 Upper CL Dif 1.16737 Prob > |t| 0.0497* Lower CL Dif 0.00073 Prob > t 0.0249* Confidence 0.95 Prob < t 0.9751 The results are shown in FIG. 10B. ER+ subpopulation Oneway Analysis of CD274 Expression By pCR ER = ERpos The results are shown in FIG. 11A. t Test YES-NO Assuming unequal variances Difference 0.25169 t Ratio 0.898709 Std Err Dif 0.28006 DF 6.542171 Upper CL Dif 0.92345 Prob > |t| 0.4007 Lower CL Dif −0.42006 Prob > t 0.2003 Confidence 0.95 Prob < t 0.7997 The results are shown in FIG. 11B. Oneway Analysis of IFNG Expression By pCR ER = ERpos The results are shown in FIG. 12A. t Test YES-NO Assuming unequal variances Difference 0.5931 t Ratio 1.501336 Std Err Dif 0.3951 DF 7.109044 Upper CL Dif 1.5244 Prob > |t| 0.1763 Lower CL Dif −0.3382 Prob > t 0.0882 Confidence 0.95 Prob < t 0.9118 The results are shown in FIG. 12B.

TABLE-US-00022 APPENDIX II Nominal Logistic Fit for pCR ER = ERneg Converged in Gradient, 5 iterations Whole Model Test Model −LogLikelihood DF ChiSquare Prob > ChiSq Difference 6.784783 6 13.56957 0.0348* Full 26.110299 Reduced 32.895082 RSquare (U) 0.2063 AICc 69.0206 BIC 79.319 Observations (or Sum Wgts) 48 Measure Training Definition Entropy RSquare 0.2063 1 − Loglike(model)/Loglike(0) Generalized RSquare 0.3301 (1 − (L(0)/L(model)){circumflex over ( )}(2/n))/(1 − L(0){circumflex over ( )}(2/n)) Mean −Log p 0.5440 Σ − Log(ρ[j])/n RMSE 0.4278 √Σ(y[j] − ρ[j]).sup.2/n Mean Abs Dev 0.3665 Σ |y[j] − ρ[j]|/n Misclassification Rate 0.2292 Σ (ρ[j] ≠ ρMax)/n N 48 n Lack Of Fit Source DF −LogLikelihood ChiSquare Lack Of Fit 41 26.110299 52.2206 Saturated 47 0.000000 Prob > ChiSq Fitted 6 26.110299  0.1125 Parameter Estimates Term Estimate Std Error ChiSquare Prob > ChiSq Lower 95% Upper 95% Intercept −5.9688255 4.1632695 2.06 0.1517 −15.115329 1.70408281 Patient Age 0.04906238 0.0425045 1.33 0.2484 −0.0324034 0.13829525 Cancer Type[IBC] −0.0943023 1.0982289 0.01 0.9316 −2.5407977 2.23824618 Cancer −0.1514945 0.6544424 0.05 0.8169 −1.5051269 1.21757158 Type[LABC] pN[N0] 0.08157636 0.4979574 0.03 0.8699 −0.8986622 1.09707358 CD274 Expression 1.64979222 0.7194762 5.26 0.0218* 0.39533833 3.2836052 IFNG Expression −1.1882978 0.5122023 5.38 0.0203* −2.3323039 −0.2889168 For log odds of NO/YES Effect Likelihood Ratio Tests Source Nparm DF L-R ChiSquare Prob > ChiSq Patient Age 1 1 1.38574446 0.2391 Cancer Type 2 2 0.19781033 0.9058 pN 1 1 0.02690704 0.8697 CD274 Expression 1 1 7.09800433 0.0077* IFNG Expression 1 1 7.15387723 0.0075* Odds Ratios For pCR odds of NO versus YES Tests and confidence intervals on odds ratios are likelihood ratio based. Unit Odds Ratios Per unit change in regressor Term Odds Ratio Lower 95% Upper 95% Reciprocal Patient Age 1.050286 0.968116 1.148315 0.9521217 CD274 Expression 5.205898 1.484886 26.67176 0.1920898 IFNG Expression 0.30474 0.097072 0.749074 3.2814908 Level1 /Level2 Odds Ratio Prob > Chisq Lower 95% Upper 95% Odds Ratios for Cancer Type LABC IBC 0.9444125 0.9722 0.0282989 35.902054 OPERABLE IBC 1.405087 0.8471 0.0357479 68.159191 OPERABLE LABC 1.4877895 0.6568 0.2518769 9.0216463 IBC LABC 1.0588593 0.9722 0.0278536 35.337072 IBC OPERABLE 0.7116997 0.8471 0.0146715 27.973694 LABC OPERABLE 0.6721381 0.6568 0.1108445 3.9701934 Odds Ratios for pN N1 N0 0.8494615 0.8697 0.1114536 6.033483 N0 N1 1.1772165 0.8697 0.1657417 8.9723459 Receiver Operating Characteristic (see FIG. 13) Using pCR = ‘YES’ to be the positive level AUC 0.79718 Confusion Matrix Actual Predicted Training NO YES NO 22  5 YES  6 15 Lift Curve (see FIG. 14) pCR . . . NO . . . YES Prediction Profiler (see FIG. 15) Nominal Logistic Fit for pCR ER = ERpos Converged in Gradient, 19 iterations Whole Model Test Model −LogLikelihood DF ChiSquare Prob > ChiSq Difference 2.400597 6 4.801193 0.5696 Full 14.343001 Reduced 16.743598 RSquare (U) 0.1434 AICc 46.2989 BIC 54.3309 Observations (or Sum Wgts) 39 Measure Training Definition Entropy RSquare 0.1434 1 − Loglike(model)/Loglike(0) Generalized RSquare 0.2010 (1 − (L(0)/L(model)){circumflex over ( )}(2/n))/(1 − L(0){circumflex over ( )}(2/n)) Mean −Log p 0.3678 Σ − Log(ρ[j])/n RMSE 0.3462 √Σ(y[j] − ρ[j]).sup.2/n Mean Abs Dev 0.2351 Σ |y[j] − ρ[j]|/n Misclassification Rate 0.1795 Σ (ρ[j] ≠ ρMax)/n N 39 n Lack Of Fit Source DF −LogLikelihood ChiSquare Lack Of Fit 32 14.343001 28.686 Saturated 38 0.000000 Prob > ChiSq Fitted 6 14.343001  0.6351 Parameter Estimates Term Estimate Std Error ChiSquare Prob > ChiSq Lower 95% Upper 95% Intercept Unstable 7.20306909 3597.5107 0.00 0.9984 −7043.7884 7058.1945 Patient Age 0.0578149 0.0628112 0.85 0.3573 −0.0560483 0.19608254 Cancer Unstable 12.0092513 7195.0139 0.00 0.9987 −14089.959 14113.9773 Type[IBC] Cancer Unstable −6.5864683 3597.507 0.00 0.9985 −7057.5706 7044.39766 Type[LABC] pN[N0] −0.0542869 0.5572904 0.01 0.9224 −1.1698378 1.117206 CD274 0.08485271 0.9859164 0.01 0.9314 −1.8704698 2.14104768 Expression IFNG −0.7334678 0.6191817 1.40 0.2362 −2.0476903 0.45985303 Expression For log odds of NO/YES Effect Likelihood Ratio Tests Source Nparm DF L-R ChiSquare Prob > ChiSq Patient Age 1 1 0.92588732 0.3359 Cancer Type 2 2 1.89140212 0.3884 pN 1 1 0.00946444 0.9225 CD274 Expression 1 1 0.00742213 0.9313 IFNG Expression 1 1 1.45693945 0.2274 Odds Ratios For pCR odds of NO versus YES Tests and confidence intervals on odds ratios are likelihood ratio based. Unit Odds Ratios Per unit change in regressor Term Odds Ratio Lower 95% Upper 95% Reciprocal Patient Age 1.059519 0.945493 1.216627 0.9438246 CD274 Expression 1.088557 0.154051 8.508347 0.9186476 IFNG Expression 0.480241 0.129033 1.583841 2.0822891 Level1 /Level2 Odds Ratio Prob > Chisq Lower 95% Upper 95% Odds Ratios for Cancer Type LABC IBC 8.3942e−9 0.2128 0 5.1523961 OPERABLE IBC 2.6876e−8 0.4499 0 20.868673 OPERABLE LABC 3.2017112 0.3193 0.2999262 36.429388 IBC LABC 119129251 0.2128 0.1940845 . IBC OPERABLE 37207993 0.4499 0.0479187 . LABC OPERABLE 0.312333 0.3193 0.0274504 3.3341535 Odds Ratios for pN N1 N0 1.1146872 0.9225 0.1070551 10.377869 N0 N1 0.8971126 0.9225 0.0963589 9.3409878 Receiver Operating Characteristic (see FIG. 16) Using pCR = ‘YES’ to be the positive level AUC 0.77273 Confusion Matrix Actual Predicted Training NO YES NO 32 1 YES  6 0

[0270] The present invention refers to the following nucleotide and amino acid sequences:

[0271] The sequences provided herein are, inter alia, available in the NCBI database and disclosed in WO 2010/077634 and can be retrieved from world wide web at ncbi.nlm.nih.gov/sites/entrez?db=gene; Theses sequences also relate to annotated and modified sequences. The present invention also provides techniques and methods wherein homologous sequences, and variants of the concise sequences provided herein are used.

[0272] SEQ ID NOS: 1-21 define the anti-PD-L1 antibody to be used in accordance with the present invention. SEQ ID NOS: 1-21 are shown in the sequence listing.

[0273] SEQ ID No. 22 to 37 show sequences of amino acid sequences for Domains I-IV of the HER2 protein (SEQ ID NO. 22-25, see also FIG. 1) and sequences of anti-HER2-antibodies. (SEQ ID NOS: 26 to 37; see also FIGS. 2A, 2B, 3A, 3B, 4A, 4B, 5A, and 5B).

SEQ ID No. 26:

Amino Acid Sequence of the Variable

[0274] light (Vr) (FIG. 2A) domain of murine monoclonal antibody 2C4 (SEQ ID NOS: 26 and 27, respectively) as shown in FIGS. 2A and 2B.

SEQ ID No. 27:

[0275] Amino acid sequence of the variable heavy (V.sub.H) (FIG. 2B) domain of murine monoclonal antibody 2C4 as shown in FIGS. 2A and 2B.

SEQ ID No. 28:

[0276] Amino acid sequence of the variable light (V.sub.L) (FIG. 2A) domain of variant 574/Pertuzumab as shown in FIGS. 2A and 2B.

SEQ ID No. 29:

[0277] Amino acid sequence of the variable heavy (V.sub.H) (FIG. 2B) domain of variant 574/Pertuzumab as shown in FIGS. 2A and 2B.

SEQ ID No. 30:

[0278] human V.sub.L consensus frameworks (hum κ1, light kappa subgroup I; humIII, heavy subgroup III) as shown in FIGS. 2A and 2B.

SEQ ID No. 31:

[0279] human V.sub.H consensus frameworks (hum κ1, light kappa subgroup I; humIII, heavy subgroup III) as shown in FIGS. 2A and 2B.

SEQ ID No. 32:

[0280] Amino acid sequences of Pertuzumab light chain as shown in FIG. 3A.

SEQ ID No. 33:

[0281] Amino acid sequences of Pertuzumab heavy chain as shown in FIG. 3B.

SEQ ID No. 34:

[0282] Amino acid sequence of Trastuzumab light chain domain as shown in FIG. 4A Boundaries of the variable light domain are indicated by arrows.

SEQ ID No. 35:

[0283] Amino acid sequence of Trastuzumab heavy chain as shown in FIG. 4B. Boundaries of the variable heavy domain are indicated by arrows.

SEQ ID No. 36:

[0284] Amino acid sequence of variant Pertuzumab light chain sequence (FIG. 5A).

SEQ ID No. 37:

[0285] Amino acid sequence of variant Pertuzumab heavy chain sequence (FIG. 5B).

SEQ ID NO. 38:

[0286] Nucleotide sequence encoding Homo sapiens Progesterone Receptor (PR)

[0287] NCBI Reference Sequence: NC_000011.9

>gi|224589802:c101000544-100900355 Homo sapiens chromosome 11, GRCh37.p10 Primary Assembly

SEQ ID No. 39:

[0288] Amino acid sequence of Homo sapiens Progesterone Receptor (PR)

[0289] PRGR_HUMAN Length: 933 Dec. 7, 2012 15:10 Type: P Check: 6067.

SEQ ID NO. 40:

[0290] Nucleotide sequence encoding Homo sapiens Estrogen Receptor (ER) (NM 000125.3)

SEQ ID NO. 41:

[0291] Nucleotide sequence encoding Homo sapiens Estrogen Receptor (ER)

[0292] NCBI Reference Sequence: NC 000006.11

>gi|224589818:152011631-152424409 Homo sapiens chromosome 6, GRCh37.p10 Primary Assembly

SEQ ID No. 42:

[0293] Amino acid sequence of Homo sapiens Estrogen Receptor (ER)

>ENST00000206249 6

SEQ ID No. 43:

[0294] Nucleotide sequence encoding Homo sapiens programmed death ligand 1 (PD-L1)

[0295] NCBI Reference Sequence: NC 000009.11

>gi|224589821:5450503-5470567 Homo sapiens chromosome 9, GRCh37.p10 Primary Assembly

SEQ ID NO. 44

[0296] Nucleotide sequence encoding Homo sapiens programmed death ligand 1(PD-L1) (CD274), transcript variant 1, mRNA

[0297] NCBI Reference Sequence: NM 014143.3

>gi|292658763|ref|NM_014143.3|Homo sapiens CD274 molecule (CD274), transcript variant 1, mRNA

SEQ ID No.45:

[0298] Amino acid sequence of Homo sapiens programmed death ligand 1(PD-L1) (programmed cell death 1 ligand 1 isoform a precursor [Homo sapiens])

[0299] NCBI Reference Sequence: NP 054862.1

>gi|7661534|ref|NP_054862.1|programmed cell death 1 ligand 1 isoform a precursor [Homo sapiens]

SEQ ID No. 46:

[0300] Nucleotide sequence encoding Homo sapiens programmed death ligand 1(PD-L1) (CD274), transcript variant 2, mRNA

[0301] NCBI Reference Sequence: NM 001267706.1

>giβ90979638|ref|NM_001267706.1|Homo sapiens CD274 molecule (CD274), transcript variant 2, mRNA

SEQ ID No. 47:

[0302] Amino acid sequence of Homo sapiens programmed death ligand 1(PD-L1) (programmed cell death 1 ligand 1 isoform b precursor [Homo sapiens])

[0303] NCBI Reference Sequence: NP 001254635.1

>gi|390979639|ref|NP_001254635.1| programmed cell death 1 ligand 1 isoform b precursor [Homo sapiens]

SEQ ID No. 48:

[0304] Nucleotide sequence encoding Homo sapiens programmed death ligand 1(PD-L1) (Homo sapiens CD274 molecule (CD274), transcript variant 3, non-coding RNA)

[0305] NCBI Reference Sequence: NR 052005.1

>gi|390979640|ref|NR_052005.1| Homo sapiens CD274 molecule (CD274), transcript variant 3, non-coding RNA

SEQ ID No. 49:

[0306] Nucleotide sequence encoding Homo sapiens interferon gamma (Homo sapiens chromosome 12, GRCh37.p10 Primary Assembly)

[0307] NCBI Reference Sequence: NC 000012.11

>gi|224589803:c68553521-68548550 Homo sapiens chromosome 12, GRCh37.p10 Primary Assembly

SEQ ID No. 50:

[0308] Nucleotide sequence encoding Homo sapiens interferon gamma, mRNA

[0309] NCBI Reference Sequence: NM 000619.2

>gi|56786137|ref|NM_000619.2|Homo sapiens interferon, gamma (IFNG), mRNA

SEQ ID No. 51:

[0310] Amino acid sequence of Homo sapiens interferon gamma, interferon gamma precursor [Homo sapiens]

[0311] NCBI Reference Sequence: NP 000610.2

>gi|56786138|ref|NP_000610.21 interferon gamma precursor [Homo sapiens]

[0312] All references cited herein are fully incorporated by reference. Having now fully described the invention, it will be understood by a person skilled in the art that the invention may be practiced within a wide and equivalent range of conditions, parameters and the like, without affecting the spirit or scope of the invention or any embodiment thereof.