DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Double Patenting
There is no non-statutory double patenting based on parent application 17/749,775 because the parent application is a divisional of a grandparent application 16/975,647, and the instant Application recites claims fairly analogous to original claims 22-30 of the grandparent application, which were withdrawn as a non-elected claim set in grandparent application. Claims analogous to these grandparent application claims 22-30 were not examined in the parent application.
Claim Objections
Applicant is advised that should claim 51 be found allowable, claim 54 will be objected to under 37 CFR 1.75 as being a substantial duplicate thereof. When two claims in an application are duplicates or else are so close in content that they both cover the same thing, despite a slight difference in wording, it is proper after allowing one claim to object to the other as being a substantial duplicate of the allowed claim. See MPEP § 608.01(m).
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-17 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. Independent claim 1 recites, “training data comprising imaging data acquired from one or more subjects and biological feature data determined from biopsies collected from the one or more subjects” and claim 10 recites, “the training data comprise biological feature data determined from biopsies collected from one or more subjects and imaging data acquired from one or more subjects, wherein the imaging data are spatially matched to locations from which the biopsies were collected in the one or more subjects”. However, given the context provided by the disclosure, “imaging data” can be reasonably interpreted as any type of medical image – including histological images, such as images of a sample on a slide captured by a microscope, for example. Similarly, another example of medical imaging data can be an optical image taken of a patient’s surface – for medical purposes. However, the specification does not show that Applicant has possession of this full scope of “imaging data”. The following paragraphs of the filed specification indicate that Applicant has possession of radiological medical imaging data:
PNG
media_image1.png
286
1112
media_image1.png
Greyscale
PNG
media_image2.png
240
1106
media_image2.png
Greyscale
PNG
media_image3.png
235
1106
media_image3.png
Greyscale
PNG
media_image4.png
191
1107
media_image4.png
Greyscale
The claimed, “biological feature data determined from biopsies” is supported in the specification as histological imaging data, such as a biopsied sampled imaged by a microscope. However, this is recited as a separate element of the two types of data used in training the machine learning model, namely, the “imaging data” and the “biological feature data determined from biopsies”. Once again, histological, microscope imaging data is only disclosed as pertaining to the “biological feature data determined from biopsies” training data limitation and not the “imaging data” training data limitation.
Depending claims do not remedy this deficiency.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Hu LS, Ning S, Eschbacher JM, Gaw N, Dueck AC, Smith KA, et al. (2015) Multi-Parametric MRI and Texture Analysis to Visualize Spatial Histologic Heterogeneity and Tumor Extent in Glioblastoma. PLoS ONE 10(11): e0141506 (Hu).
As per claim 1, Hu teaches a method for constructing and implementing a machine learning model to generate at least one image that depicts spatial patterns of a biomarker across a region-of-interest in a subject, the method comprising:
constructing a trained machine learning model by:
(i) accessing training data with a computer system, the training data comprising imaging data acquired from one or more subjects and biological feature data determined from biopsies collected from the one or more subjects (Hu: abstract: “We recruited primary GBM patients undergoing image-guided biopsies and acquired preoperative MRI: CE-MRI, Dynamic-Susceptibility-weighted-Contrast-enhanced-MRI, and Diffusion Tensor Imaging”;
Page 3, para 4 – page 4, last para: “Surgical biopsy”, “Histologic analysis and biopsy sample classification”, “MRI protocol, parametric maps, and image coregistration”);
(ii) quantifying regional biomarker diversity in the one or more subjects from the biological feature data (Hu: abstract: “Following image coregistration and region of interest placement at biopsy locations, we compared MRI metrics and regional texture with histologic diagnoses of high- vs low-tumor content (>/= 80% vs <80% tumor nuclei) for corresponding samples.”
Introduction: “The Cancer Genome Atlas (TCGA) have sought to catalog GBM’s diverse genetic landscape, giving insight to pathogenesis, prognosis and therapeutic susceptibility. This should help guide risk stratification for existing protocols and help identify key driver genes as potential therapeutic targets in the future”;
Page 3, last para – page 4, 1st para:
PNG
media_image5.png
509
1203
media_image5.png
Greyscale
Page 6, last para – page 7, 1st para: “TCGA”
PNG
media_image6.png
404
1203
media_image6.png
Greyscale
PNG
media_image7.png
704
1654
media_image7.png
Greyscale
Page 11, para 2: “In this study, we define high tumor content (>/= 80% tumor) based on published TCGA criteria for sample adequacy [4]. This threshold helps to maximize tumoral DNA quality by minimizing non-tumoral contamination”);
(iii) training a machine learning model based on the training data and the quantified regional biomarker diversity in the one or more subjects, wherein the machine learning model is trained on the training data to localize distinct subpopulations across a region-of-interest (Hu: abstract: “Following image coregistration and region of interest placement at biopsy locations, we compared MRI metrics and regional texture with histologic diagnoses of high- vs low-tumor content (>/= 80% vs <80% tumor nuclei) for corresponding samples. In a training set, we used three texture analysis algorithms and three ML methods to identify MRI-texture features that optimized model accuracy to distinguish tumor content…
The MRI-based model achieved 85% cross-validated accuracy to diagnose high- vs low-tumor in the training set (60 biopsies, 11 patients)”;
Page 3, para 2 (shown below): “using machine learning (ML) to integrate an array of image-based
texture features from pre-operative MRI to predict tumor-rich biopsies… develop non-invasive correlates of histology that can facilitate image-guided biopsy and genomic profiling”
Page 6, para 2:
PNG
media_image8.png
430
1209
media_image8.png
Greyscale
Page 3, para 2 (shown below): “We used classification algorithms with sequential forward feature selection to identify the subset of image-based PCs (determined from PCA above) with the greatest combined discrimination for biopsy tumor content (high- vs low-tumor) …
With LOOCV, all samples but one are used to develop the classifier (i.e., training set) with one randomly chosen sample serving as the test case for classification performance [13,14]. This process repeats for all samples in the dataset (i.e., 60 separate trial runs in our cohort), until each sample in the cohort has served as the test sample. The averaged accuracy is the overall cross validation (CV) accuracy. In building the classification model, the PCs are selected using sequential forward feature selection”;
Page 8, para 1: “Ten percent of training set biopsies (n = 6) were located within 5-10mm of adjacent biopsies”:
Fig. 1: “ML analysis and multi-parametric MRI in 60 training biopsies”;
Page 9, last para: “Based on the correlations in our training set, we have used machine learning (ML) to identify three MRI-based features that optimize classification accuracy for high- vs. low-tumor content (Table 2)”:
PNG
media_image9.png
295
1656
media_image9.png
Greyscale
Page 11, para 2: “In this study, we define high tumor content (>/= 80% tumor) based on published TCGA criteria for sample adequacy [4]. This threshold helps to maximize tumoral DNA quality by minimizing non-tumoral contamination”); and
generating an image that depicts spatial patterns of a biomarker across a region-of-interest in a subject by inputting images acquired from the subject to the trained machine learning model (Hu: title: “Multi-Parametric MRI and Texture Analysis to Visualize Spatial Histologic Heterogeneity and Tumor Extent in Glioblastoma”;
abstract: “In this study, we use multiple texture analysis and machine learning
(ML) algorithms to analyze multi-parametric MRI, and produce new images indicating
tumor-rich targets in GBM…
Multi-parametric MRI and texture analysis can help characterize and visualize GBM’s spatial
histologic heterogeneity to identify regional tumor-rich biopsy targets.”;
Page 3, para 2:
PNG
media_image10.png
472
1734
media_image10.png
Greyscale
Page 6, para 3:
PNG
media_image11.png
679
1206
media_image11.png
Greyscale
Page 8, para 1: “using the MRI texture model would significantly improve the localization and
recovery of tumor-rich BAT targets compared to current CE-MRI guided biopsy methods, as
shown in Fig 1. Ten percent of training set biopsies (n = 6) were located within 5-10mm of
adjacent biopsies”:
PNG
media_image12.png
919
1236
media_image12.png
Greyscale
: “Fig 1. ML-based MRI invasion maps show tumor-rich (>80% tumor nuclei) extent throughout ENH and BAT… Color overlay maps with manual tracings (green) around BAT show the probability (range 0–1) of tumor-rich (red) vs tumor-poor (green/blue) content, based on ML analysis and multi-parametric MRI in 60 training biopsies and 22 validation biopsies. The maps show correspondence between tumor-rich (B, red) and tumor-poor (D, blue/gray) biopsy samples”;
Page 9, para 1:
PNG
media_image13.png
715
1203
media_image13.png
Greyscale
Page 11, para 1: “Using the ML-based model to map tumor infiltration in BAT”;
Page 11, para 2: “In this study, we define high tumor content (>/= 80% tumor) based on published TCGA criteria for sample adequacy [4]. This threshold helps to maximize tumoral DNA quality by minimizing non-tumoral contamination”).
As per claim 2, Hu teaches the method of claim 1, wherein the biomarker comprises a histological feature (Hu: See arguments and citations offered in rejecting claim 1 above: histolog*).
As per claim 3, Hu teaches the method of claim 2, wherein the histological feature includes cell density (Hu: See arguments and citations offered in rejecting claim 2 above;
page 2, last para: “Image-based features such as tumor cell density on diffusion-weighted imaging (DWI), white matter infiltration on diffusion tensor imaging (DTI), and microvessel morphology on perfusion MRI (pMRI) reflect key biophysical characteristics associated with tumor pathogenesis”;
Page 11, para 3: “additional ML models can be developed in future work for other clinically relevant histologic thresholds, or potentially for prediction of tumor cell density as a continuous variable (i.e., tumor nuclei range from 0–100%). We must also note that different histologic thresholds may favor specific MRI based phenotypes. For instance, high tumor density may select for “proliferative” imaging phenotypes, while lower tumor thresholds (i.e., 50% or 20% tumor fraction) may associate with “invasive” phenotypes in which tumor admixes more evenly with surrounding nontumoral parenchyma”).
As per claim 4, Hu teaches the method of claim 1, wherein the biomarker includes a biological descriptor of a tumor comprising at least one of tumor aggressiveness, tumor subtype, or likelihood of recurrence (Hu: See arguments and citations offered in rejecting claim 1 above;
page 2, para 2: “We collected 82 biopsies from 18 GBMs throughout ENH and BAT. The MRI-based model achieved 85% cross-validated accuracy to diagnose high- vs low-tumor in the training set (60 biopsies, 11 patients)”).
As per claim 5, Hu teaches the method of claim 1, wherein the biomarker indicates interactions between the distinct subpopulations across the region-of-interest (Hu: See arguments and citations offered in rejecting claim 1 above: Cell interactions are the dynamic state of different cells in the tissue region of interest – including growth.
ENH and BAT;
Para 4, para 1: “Taking into account all visible cells (neurons, inflammatory cells, reactive glia, tumor cells, etc.), the percent tumor nuclei were estimated. Based on the TCGA criteria, we used the threshold of at least 80% tumor nuclei content to define tumor-rich (i.e., high tumor) biopsy samples [4]. Those with less than 80% tumor content were classified as low tumor samples”).
As per claim 6, Hu teaches the method of claim 5, wherein the interactions between the distinct subpopulations comprise cell interactions (Hu: See arguments and citations offered in rejecting claim 5 above).
As per claim 7, Hu teaches the method of claim 6, wherein the distinct subpopulations comprise molecularly distinct subpopulations (Hu: See arguments and citations offered in rejecting claim 5 above).
As per claim 8, Hu teaches the method of claim 1, wherein the trained machine learning model is a transfer learning model trained on imaging data and biological feature data acquired from the subject (Hu: See arguments and citations offered in rejecting claim 1 above: training data is from patients used in validation and testing;
Page 6, para 2: “With LOOCV, all samples but one are used to develop the classifier (i.e., training set) with one randomly chosen sample serving as the test case for classification performance [13,14]. This process repeats for all samples in the dataset (i.e., 60 separate trial runs in our cohort), until each
sample in the cohort has served as the test sample”).
As per claim 9, Hu teaches the method of claim 1, wherein quantifying regional biomarker diversity in the one or more subjects from the biological feature data comprises quantifying the biomarker diversity based on one or more continuous variables (Hu: See arguments and citations offered in rejecting claim 1 above).
As per claim(s) 10-12 and 14-17, arguments made in rejecting claim(s) 1-3, 5-7, and 9 are analogous, respectively.
As per claim 13, Hu teaches the method of claim 1, wherein the biomarker includes a biological descriptor of a tumor comprising at least one of tumor aggressiveness, tumor subtype, or likelihood of recurrence, such that the trained machine learning model is trained to generate images that depict spatial patterns of the biological description of the tumor (Hu: See arguments and citations offered in rejecting claims 1 and 4 above).
As per claim(s) 18, arguments made in rejecting claim(s) 1 are analogous, respectively.
As per claim 19, Hu teaches the method of claim 18, wherein the training data further comprise regional biomarker diversity data generated by quantifying regional biomarker diversity in the plurality of subjects from the biological feature data (Hu: See arguments and citations offered in rejecting claim 1 above).
As per claim 20, Hu teaches the method of claim 18, wherein the biomarker includes a biological descriptor of a tumor comprising at least one of tumor aggressiveness, tumor subtype, or likelihood of recurrence (Hu: See arguments and citations offered in rejecting claims 1, 4, 13 above).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Atiba Fitzpatrick whose telephone number is (571) 270-5255. The examiner can normally be reached on M-F 10:00am-6pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Bee can be reached on (571) 270-5183. The fax phone number for Atiba Fitzpatrick is (571) 270-6255.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Atiba Fitzpatrick
/ATIBA O FITZPATRICK/
Primary Examiner, Art Unit 2677
Prosecution Insights