Mass Spectrometry-Based Multivariate Proteomic Tests for Prediction of Outcomes on Immune Checkpoint Blockade Therapy: The Modern Analytical Approach.

The remarkable success of immune checkpoint inhibitors (ICIs) has given hope of cure for some patients with advanced cancer; however, the fraction of responding patients is 15-35%, depending on tumor type, and the proportion of durable responses is even smaller. Identification of biomarkers with strong predictive potential remains a priority. Until now most of the efforts were focused on biomarkers associated with the assumed mechanism of action of ICIs, such as levels of expression of programmed death-ligand 1 (PD-L1) and mutation load in tumor tissue, as a proxy of immunogenicity; however, their performance is unsatisfactory. Several assays designed to capture the complexity of the disease by measuring the immune response in tumor microenvironment show promise but still need validation in independent studies. The circulating proteome contains an additional layer of information characterizing tumor-host interactions that can be integrated into multivariate tests using modern machine learning techniques. Here we describe several validated serum-based proteomic tests and their utility in the context of ICIs. We discuss test performances, demonstrate their independence from currently used biomarkers, and discuss various aspects of associated biological mechanisms. We propose that serum-based multivariate proteomic tests add a missing piece to the puzzle of predicting benefit from ICIs.

Extending the information content of the MALDI analysis of biological fluids via multi-million shot analysis.

INTRODUCTION: Reliable measurements of the protein content of biological fluids like serum or plasma can provide valuable input for the development of personalized medicine tests. Standard MALDI analysis typically only shows high abundance proteins, which limits its utility for test development. It also exhibits reproducibility issues with respect to quantitative measurements. In this paper we show how the sensitivity of MALDI profiling of intact proteins in unfractionated human serum can be substantially increased by exposing a sample to many more laser shots than are commonly used. Analytical reproducibility is also improved. METHODS: To assess what is theoretically achievable we utilized spectra from the same samples obtained over many years and combined them to generate MALDI spectral averages of up to 100,000,000 shots for a single sample, and up to 8,000,000 shots for a set of 40 different serum samples. Spectral attributes, such as number of peaks and spectral noise of such averaged spectra were investigated together with analytical reproducibility as a function of the number of shots. We confirmed that results were similar on MALDI instruments from different manufacturers. RESULTS: We observed an expected decrease of noise, roughly proportional to the square root of the number of shots, over the whole investigated range of the number of shots (5 orders of magnitude), resulting in an increase in the number of reliably detected peaks. The reproducibility of the amplitude of these peaks, measured by CV and concordance analysis also improves with very similar dependence on shot number, reaching median CVs below 2% for shot numbers > 4 million. Measures of analytical information content and association with biological processes increase with increasing number of shots. CONCLUSIONS: We demonstrate that substantially increasing the number of laser shots in a MALDI-TOF analysis leads to more informative and reliable data on the protein content of unfractionated serum. This approach has already been used in the development of clinical tests in oncology.

A dropout-regularized classifier development approach optimized for precision medicine test discovery from omics data.

BACKGROUND: Modern genomic and proteomic profiling methods produce large amounts of data from tissue and blood-based samples that are of potential utility for improving patient care. However, the design of precision medicine tests for unmet clinical needs from this information in the small cohorts available for test discovery remains a challenging task. Obtaining reliable performance assessments at the earliest stages of test development can also be problematic. We describe a novel approach to classifier development designed to create clinically useful tests together with reliable estimates of their performance. The method incorporates elements of traditional and modern machine learning to facilitate the use of cohorts where the number of samples is less than the number of measured patient attributes. It is based on a hierarchy of classification and information abstraction and combines boosting, bagging, and strong dropout regularization. RESULTS: We apply this dropout-regularized combination approach to two clinical problems in oncology using mRNA expression and associated clinical data and compare performance with other methods of classifier generation, including Random Forest. Performance of the new method is similar to or better than the Random Forest in the two classification tasks used for comparison. The dropout-regularized combination method also generates an effective classifier in a classification task with a known confounding variable. Most importantly, it provides a reliable estimate of test performance from a relatively small development set of samples. CONCLUSIONS: The flexible dropout-regularized combination approach is able to produce tests tailored to particular clinical questions and mitigate known confounding effects. It allows the design of molecular diagnostic tests addressing particular clinical questions together with reliable assessment of whether test performance is likely to be fit-for-purpose in independent validation at the earliest stages of development.

Robust identification of molecular phenotypes using semi-supervised learning.

BACKGROUND: Modern molecular profiling techniques are yielding vast amounts of data from patient samples that could be utilized with machine learning methods to provide important biological insights and improvements in patient outcomes. Unsupervised methods have been successfully used to identify molecularly-defined disease subtypes. However, these approaches do not take advantage of potential additional clinical outcome information. Supervised methods can be implemented when training classes are apparent (e.g., responders or non-responders to treatment). However, training classes can be difficult to define when assessing relative benefit of one therapy over another using gold standard clinical endpoints, since it is often not clear how much benefit each individual patient receives. RESULTS: We introduce an iterative approach to binary classification tasks based on the simultaneous refinement of training class labels and classifiers towards self-consistency. As training labels are refined during the process, the method is well suited to cases where training class definitions are not obvious or noisy. Clinical data, including time-to-event endpoints, can be incorporated into the approach to enable the iterative refinement to identify molecular phenotypes associated with a particular clinical variable. Using synthetic data, we show how this approach can be used to increase the accuracy of identification of outcome-related phenotypes and their associated molecular attributes. Further, we demonstrate that the advantages of the method persist in real world genomic datasets, allowing the reliable identification of molecular phenotypes and estimation of their association with outcome that generalizes to validation datasets. We show that at convergence of the iterative refinement, there is a consistent incorporation of the molecular data into the classifier yielding the molecular phenotype and that this allows a robust identification of associated attributes and the underlying biological processes. CONCLUSIONS: The consistent incorporation of the structure of the molecular data into the classifier helps to minimize overfitting and facilitates not only good generalization of classification and molecular phenotypes, but also reliable identification of biologically relevant features and elucidation of underlying biological processes.

Predictive value of a proteomic signature in patients with non-small-cell lung cancer treated with second-line erlotinib or chemotherapy (PROSE): a biomarker-stratified, randomised phase 3 trial.

BACKGROUND: An established multivariate serum protein test can be used to classify patients according to whether they are likely to have a good or poor outcome after treatment with EGFR tyrosine-kinase inhibitors. We assessed the predictive power of this test in the comparison of erlotinib and chemotherapy in patients with non-small-cell lung cancer. METHODS: From Feb 26, 2008, to April 11, 2012, patients (aged ≥18 years) with histologically or cytologically confirmed, second-line, stage IIIB or IV non-small-cell lung cancer were enrolled in 14 centres in Italy. Patients were stratified according to a minimisation algorithm by Eastern Cooperative Oncology Group performance status, smoking history, centre, and masked pretreatment serum protein test classification, and randomly assigned centrally in a 1:1 ratio to receive erlotinib (150 mg/day, orally) or chemotherapy (pemetrexed 500 mg/m(2), intravenously, every 21 days, or docetaxel 75 mg/m(2), intravenously, every 21 days). The proteomic test classification was masked for patients and investigators who gave treatments, and treatment allocation was masked for investigators who generated the proteomic classification. The primary endpoint was overall survival and the primary hypothesis was the existence of a significant interaction between the serum protein test classification and treatment. Analyses were done on the per-protocol population. This trial is registered with ClinicalTrials.gov, number NCT00989690. FINDINGS: 142 patients were randomly assigned to chemotherapy and 143 to erlotinib, and 129 (91%) and 134 (94%), respectively, were included in the per-protocol analysis. 88 (68%) patients in the chemotherapy group and 96 (72%) in the erlotinib group had a proteomic test classification of good. Median overall survival was 9·0 months (95% CI 6·8-10·9) in the chemotherapy group and 7·7 months (5·9-10·4) in the erlotinib group. We noted a significant interaction between treatment and proteomic classification (pinteraction=0·017 when adjusted for stratification factors; pinteraction=0·031 when unadjusted for stratification factors). Patients with a proteomic test classification of poor had worse survival on erlotinib than on chemotherapy (hazard ratio 1·72 [95% CI 1·08-2·74], p=0·022). There was no significant difference in overall survival between treatments for patients with a proteomic test classification of good (adjusted HR 1·06 [0·77-1·46], p=0·714). In the group of patients who received chemotherapy, the most common grade 3 or 4 toxic effect was neutropenia (19 [15%] vs one [<1%] in the erlotinib group), whereas skin toxicity (one [<1%] vs 22 [16%]) was the most frequent in the erlotinib group. INTERPRETATION: Our findings indicate that serum protein test status is predictive of differential benefit in overall survival for erlotinib versus chemotherapy in the second-line setting. Patients classified as likely to have a poor outcome have better outcomes on chemotherapy than on erlotinib. FUNDING: Italian Ministry of Health, Italian Association of Cancer Research, and Biodesix.

Changes in plasma mass-spectral profile in course of treatment of non-small cell lung cancer patients with epidermal growth factor receptor tyrosine kinase inhibitors.

INTRODUCTION: Our previous study showed that pretreatment serum or plasma Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry may predict clinical outcome of non-small cell lung cancer (NSCLC) patients treated with epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs). In this study, plasma proteomic profiles of NSCLC patients were evaluated in the course of EGFR TKIs therapy. MATERIALS AND METHODS: Plasma samples were collected at baseline, in the course of gefitinib therapy and at treatment withdrawal. Samples were analyzed by Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry. Acquired spectra were classified by the VeriStrat test into "good" and "poor" profiles. The association between VeriStrat classification and progression-free survival (PFS) and overall survival (OS), and types of clinical progression, was analyzed. RESULTS: Plasma samples from 111 NSCLC patients treated with gefitinib were processed. VeriStrat "good" classification at baseline correlated with longer PFS (hazard ratio [HR], 0.54; 95% confidence interval, 0.35-0.83; p = 0.005) and OS (HR, 0.40; 95% confidence interval, 0.26-0.61; p < 0.0001), when compared with VeriStrat "poor." Multivariate analysis confirmed longer PFS (HR, 0.52; p = 0.025) and OS (HR, 0.44; p = 0.001) in patients classified as VeriStrat "good", when VeriStrat was considered as a time-dependent variable. About one-third of baseline "good" classifications had changed to "poor" at the time of treatment withdrawal; progression in these patients was associated with the development of new lesions. CONCLUSIONS: Our findings support the role of VeriStrat in the assistance in treatment selection of NSCLC patients for EGFR TKI therapy and its potential utility in treatment monitoring.

Detection of tumor epidermal growth factor receptor pathway dependence by serum mass spectrometry in cancer patients.

BACKGROUND: We hypothesized that a serum proteomic profile predictive of survival benefit in non-small cell lung cancer patients treated with epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKI) reflects tumor EGFR dependency regardless of site of origin or class of therapeutic agent. METHODS: Pretreatment serum or plasma from 230 patients treated with cetuximab, EGFR-TKIs, or chemotherapy for recurrent/metastatic head and neck squamous cell carcinoma (HNSCC) or colorectal cancer (CRC) were analyzed by mass spectrometry. Each sample was classified into "good" or "poor" groups using VeriStrat, and survival analyses of each cohort were done based on this classification. For the CRC cohort, this classification was correlated with the tumor EGFR ligand levels and KRAS mutation status. RESULTS: In the EGFR inhibitor-treated cohorts, the classification predicted survival (HNSCC: gefitinib, P = 0.007 and erlotinib/bevacizumab, P = 0.02; CRC: cetuximab, P = 0.0065) whereas the chemotherapy cohort showed no survival difference. For CRC patients, tumor EGFR ligand RNA levels were significantly associated with the proteomic classification, and combined KRAS and proteomic classification provided improved survival classification. CONCLUSIONS: Serum proteomic profiling can detect clinically significant tumor dependence on the EGFR pathway in non-small cell lung cancer, HNSCC, and CRC patients treated with either EGFR-TKIs or cetuximab. This classification is correlated with tumor EGFR ligand levels and provides a clinically practical way to identify patients with diverse cancer types most likely to benefit from EGFR inhibitors. Prospective studies are necessary to confirm these findings.

VeriStrat classifier for survival and time to progression in non-small cell lung cancer (NSCLC) patients treated with erlotinib and bevacizumab.

We applied an established and commercially available serum proteomic classifier for survival after treatment with erlotinib (VeriStrat) in a blinded manner to pretreatment sera obtained from recurrent advanced NSCLC patients before treatment with the combination of erlotinib plus bevacizumab. We found that VeriStrat could classify these patients into two groups with significantly better or worse outcomes and may enable rational selection of patients more likely to benefit from this costly and potentially toxic regimen.

Mass spectrometry to classify non-small-cell lung cancer patients for clinical outcome after treatment with epidermal growth factor receptor tyrosine kinase inhibitors: a multicohort cross-institutional study.

BACKGROUND: Some but not all patients with non-small-cell lung cancer (NSCLC) respond to treatment with epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs). We developed and tested the ability of a predictive algorithm based on matrix-assisted laser desorption ionization (MALDI) mass spectrometry (MS) analysis of pretreatment serum to identify patients who are likely to benefit from treatment with EGFR TKIs. METHODS: Serum collected from NSCLC patients before treatment with gefitinib or erlotinib were analyzed by MALDI MS. Spectra were acquired independently at two institutions. An algorithm to predict outcomes after treatment with EGFR TKIs was developed from a training set of 139 patients from three cohorts. The algorithm was then tested in two independent validation cohorts of 67 and 96 patients who were treated with gefitinib and erlotinib, respectively, and in three control cohorts of patients who were not treated with EGFR TKIs. The clinical outcomes of survival and time to progression were analyzed. RESULTS: An algorithm based on eight distinct m/z features was developed based on outcomes after EGFR TKI therapy in training set patients. Classifications based on spectra acquired at the two institutions had a concordance of 97.1%. For both validation cohorts, the classifier identified patients who showed improved outcomes after EGFR TKI treatment. In one cohort, median survival of patients in the predicted "good" and "poor" groups was 207 and 92 days, respectively (hazard ratio [HR] of death in the good versus poor groups = 0.50, 95% confidence interval [CI] = 0.24 to 0.78). In the other cohort, median survivals were 306 versus 107 days (HR = 0.41, 95% CI = 0.17 to 0.63). The classifier did not predict outcomes in patients who did not receive EGFR TKI treatment. CONCLUSION: This MALDI MS algorithm was not merely prognostic but could classify NSCLC patients for good or poor outcomes after treatment with EGFR TKIs. This algorithm may thus assist in the pretreatment selection of appropriate subgroups of NSCLC patients for treatment with EGFR TKIs.