Recommendations from the Isbtc-SITC/FDA/NCI Workshop on Immunotherapy Biomarkers

Published OnlineFirst May 10, 2011; DOI: 10.1158/1078-0432.CCR-10-2234

Clinical Cancer Special Report Research

Recommendations from the iSBTc-SITC/FDA/NCI Workshop on Immunotherapy Biomarkers

Lisa H. Butterfield1, A. Karolina Palucka4,5, Cedrik M. Britten23,24, Madhav V. Dhodapkar6, Leif Hakansson28, Sylvia Janetzki7, Yutaka Kawakami29, Thomas-Oliver Kleen9, Peter P. Lee10, Cristina Maccalli31, Holden T. Maecker11, Vernon C. Maino12, Michele Maio33, Anatoli Malyguine13, Giuseppe Masucci27, Graham Pawelec25, Douglas M. Potter2, Licia Rivoltini32, Lupe G. Salazar19, Dolores J. Schendel26, Craig L. Slingluff, Jr.20, Wenru Song21, David F. Stroncek15, Hideaki Tahara30, Magdalena Thurin18, Giorgio Trinchieri14, Sjoerd H. van Der Burg34, Theresa L. Whiteside3, Jon M. Wigginton8, Francesco Marincola16, Samir Khleif17, Bernard A. Fox22, and Mary L. Disis19

Abstract Purpose: To facilitate development of innovative immunotherapy approaches, especially for treatment concepts exploiting the potential benefits of personalized therapy, there is a need to develop and validate tools to identify patients who can benefit from immunotherapy. Despite substantial effort, we do not yet know which parameters of antitumor immunity to measure and which assays are optimal for those measurements. Experimental Design: The iSBTc-SITC (International Society for Biological Therapy of Cancer-Society for Immunotherapy of Cancer), FDA (Food and Drug Administration), and NCI (National Cancer Institute) partnered to address these issues for immunotherapy of cancer. Here, we review the major challenges, give examples of approaches and solutions, and present our recommendations. Results and Conclusions: Although specific immune parameters and assays are not yet validated, we recommend following standardized (accurate, precise, and reproducible) protocols and use of functional assays for the primary immunologic readouts of a trial; consideration of central laboratories for immune monitoring of large, multi-institutional trials; and standardized testing of several phenotypic and functional potential potency assays specific to any cellular product. When reporting results, the full QA (quality assessment)/QC (quality control) should be conducted and selected examples of truly representative raw data and assay performance characteristics should be included. Finally, to promote broader analysis of multiple aspects of immunity, and gather data on variability, we recommend that in addition to cells and serum, RNA and DNA samples be banked (under standardized conditions) for later testing. We also recommend that sufficient blood be drawn to allow for planned testing of the primary

Authors' Affiliations: 1Departments of Medicine, Surgery and Immunology, versity of Tuebingen Medical School, Tuebingen, 26Institute of Molecular University of Pittsburgh, 2Biostatistics Department, Graduate School of Immunology and Immune Monitoring Group, Helmholth Zentrum, Public Health, University of Pittsburgh and Biostatistics Facility, 3Depart- Munchen, Germany; 27Department of Oncology-Pathology, Karolinska ment of Pathology, University of Pittsburgh Cancer Institute, Pittsburgh, Institutet/and University Hospital Coordinator of the NCEV (Nordic net- Pennsylvania; 4Baylor Institute for Immunology Research, Dallas, Texas; work Centrum of Excellence for antitumour Vaccination), Stockholm, 5Department of Gene and Cell Medicine, Immunology Institute, Mount Sinai 28CamImGuide Therapeutics AB, Hollviken, Sweden; 29Division of Cellular School of Medicine, New York, New York; 6Department of Hematology & Signaling, Institute for Advanced Medical Research, Keio University Immunobiology, Yale University, New Haven, Connecticut; 7ZellNet Con- School of Medicine, Shinjuku, 30Department of Surgery and Bioengineer- sulting, Inc., Fort Lee, 8Discovery Medicine-Clinical Oncology, Bristol-Myers ing, Advanced Clinical Research Center, Institute of Medical Science, The Squibb, Inc., Princeton, New Jersey; 9Cellular Technology Limited, Shaker University of Tokyo, Minato-ku, Tokyo, Japan; 31Unit of Immuno-biother- Heights, Ohio; 10Department of Medicine (Hematology), Stanford University apy of Melanoma and Solid Tumors, Division of Molecular Oncology, San School of Medicine, 11Human Immune Monitoring Center, Institute for Raffaele Scientific Institute, 32Unit of Immunotherapy of Human Tumors, Immunity, Transplantation, and Infection, Stanford University Medical Fondazione IRCCS Istituto Nazionale Tumori, Milan, 33Medical Oncology School, Stanford, 12BD Biosciences, San Jose, California; 13Applied and Immunotherapy, University Hospital of Siena, Istituto Toscano and Developmental Research Directorate, SAIC-Frederick, Inc., 14Cancer Tumori, Siena and Cancer Bioimmunotherapy Unit, Centro di Riferimento and Inflammation Program, CCR, National Cancer Institute (NCI), NIH, Oncologico, Aviano, Italy; 34Department of Clinical Oncology, Leiden Frederick, 15Department of Transfusion Medicine, Clinical Center, University Medical Center, Leiden, the Netherlands 16Department of Translational Medicine, 17Cancer Vaccine Section, NCI, NIH, Bethesda, 18Cancer Diagnosis Program, Division of Cancer Corresponding Author: Lisa H. Butterfield, Departments of Medicine, Treatment and Diagnosis, NCI, NIH, Rockville, Maryland; 19Tumor Vaccine Surgery and Immunology, University of Pittsburgh, 1.27 Hillman Cancer Group, Division of Oncology, University of Washington, Seattle, Washing- Center 5117 Centre Avenue, Pittsburgh, PA 15213. Phone: 412-623-1418; ton; 20Department of Surgery, University of Virginia, Charlottesville, Fax: 412-624-0264; E-mail: [email protected] or Mary L. Disis, Tumor Virginia; 21Pfizer Oncology, New London, Connecticut; 22Laboratory of Vaccine Group, Center for Translational Medicine in Women's Health, 815 Molecular and Tumor Immunology, Earle A. Chiles Research Institute, Mercer Street, University of Washington, Seattle, WA 98109. Phone: 206- Robert W. Franz Cancer Center, Providence Portland Medical Center, 616-1823; Fax: 206-685-3128; E-mail: [email protected] 23 Portland, Oregon; Department of Internal Medicine, University Medical doi: 10.1158/1078-0432.CCR-10-2234 Center of the Johannes Gutenberg-University, 24Clinical Development, BioNTech AG, Mainz, 25Second Department of Internal Medicine, Uni- 2011 American Association for Cancer Research.

3064 Clin Cancer Res; 17(10) May 15, 2011

Immunotherapy Biomarkers Recommendations

hypothesis being addressed in the trial, and that additional baseline and posttreatment blood is banked for testing novel hypotheses (or generating new hypotheses) that arise in the field. Clin Cancer Res; 17(10); 3064–76. 2011 AACR.

* A lack of agreement over which immunologic para- Translational Relevance meter to measure, what time point(s) is relevant, and which assay to use to make that measurement. Progress in the field of immunotherapy has been * Immunologic assays used may not accurately reflect slow, but recent clinical successes have given strong an in vivo immune response, for example cytotoxicity support to the potential of this approach as a treatment assays with lengthy in vitro stimulation. modality in cancer. To define biomarkers to identify * The analysis of single parameters alone may not patients who will have clinical benefit, and to ultimately provide sufficient insights about complex immune identify appropriate patient groups to enroll, clinical system–tumor interactions. trials testing immune-based interventions should con- * Common immunoassays do not take into account duct more thorough and standardized immunologic changes: in the differentiation of immune cells, in the assays to fully study clinical responders and nonrespon- antigenic profile of tumors and responding T cells, in ders, and report data and their analysis methodology in T-cell homing receptors, or the complex analysis of greater detail. An increased focus on immune assess- "private" antigen responses. ments in these patients will allow us to learn more about the mechanisms of action of the tested interven- Several Key Issues Must Be Addressed to Move tions and the positive and negative immune responses the Field Forward in treated patients. * How to ensure the quality of sample processing and storage for the future studies? * Immunotherapy of cancer is becoming increasingly How to standardize the assays and which assays utilized and is becoming a part of cancer management should we standardize? * alongside more classical approaches such as chemotherapy How to link immune and clinical outcomes with or radiotherapy. The U.S. Food and Drug Administration cellular product characteristics? * (FDA) has recently granted full approval for the first ther- How to utilize the existing knowledge and assays to apeutic cancer vaccine "Provenge" (1) on the basis of a allow for meta-analyses of laboratory and clinical randomized phase III trial (2), and other recent trials have results? reached their primary endpoints according to the investi- Following discussions between representatives of the gators (3–5). However, challenges exist that need to be FDA leadership, the National Cancer Institute (NCI), the resolved to facilitate development of innovative approa- NIH and Industry, as well as experts in the field of immu- ches, especially for treatment concepts exploiting the notherapy, the iSBTc-SITC/NCI/FDA Taskforce on immu- potential benefits of personalized therapy. There is a need notherapy biomarkers was created. This group led a for development and validation of tools to identify patients workshop held on October 28, 2009, in conjunction with who can benefit from a particular form of immunotherapy. the Annual Meeting of the International Society for Biolo- For example, only a fraction of patients are eligible for gical Therapy of Cancer; iSBTc (now known as the Society adoptive tumor-infiltrating lymphocyte (TIL) cell transfer for Immunotherapy of Cancer, SITC). This workshop was (6), only a fraction of patients can achieve durable regres- also a follow-up to the 2001 workshop of the iSBTc-SITC sions in response to cell or antigen vaccination (7), or (12) and included participation from 6 partner societies antibody therapies, and we do not know the mechanisms and representation of 20 countries. The results of the dis- responsible. Despite substantial efforts from many groups, cussions of the taskforce and recommendations from the we do not know which parameters of immune responses, workshop participants and from representatives of several and which assays used to assess these parameters, are international immunotherapy societies are presented here. optimal for efficacy analysis. Indeed, the tumor-specific cellular immune response promoted by immunization Background often has not correlated with clinical cancer regression despite the induced cytotoxic T cells detected in in vitro The major goal for the field of immunotherapy is to assays (8–11). The major reason is that objective clinical improve the clinical efficacy of immunotherapies. To do so, response rates are usually below 10%, preventing mean- we require immunologic biomarkers of efficacy. The NCI ingful correlations of specific T-cell response rates with Translational Research Working Group (TRWG) has incor- clinical responses in small sized, early stage trials. Other porated the need for these biomarkers in their develop- possible reasons are: mental pathways, which are frameworks for bench to

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3065

Butterfield et al.

bedside development of new therapies and which include a Challenges, examples, and recommendations pathway for "Immune Response Modifiers" (13). The FDA Processing and storage of blood samples to bank PBMCs pharmacogenomics guidance (14) defines a valid biomar- and serum for immunologic studies. Challenge. Multiple ker as "a biomarker that is measured in an analytic test variables in drawing blood and obtaining tissue can impact system with well-established performance characteristics the quality of the cells obtained and their proper testing. and for which there is an established scientific framework Examples. This area has been addressed in depth by the or body of evidence that elucidates the physiologic, tox- Immunologic Monitoring Consortium (investigators from icologic, pharmacologic, or clinical significance of the test the University of Washington, Duke University, BD Bios- results." This clearly states the need for biomarker assay ciences, and Coulter), and their cryopreservation and thaw- standardization and also implies the inherent complexity ing recommendations are outlined in Appendix 1A. In in accomplishing this goal in the field of immunotherapy addition, recent studies have tested the importance of time of cancer. from blood draw to PBMC processing and established that The "Critical Path" is the FDA’s initiative to identify and the shorter the time, the superior the viability and func- prioritize the most pressing medical product development tionality of the PBMCs (17, 18). The use of cell preparation problems and the greatest opportunities for rapid improve- tubes (CPT) versus Ficoll for PBMC separation, and ship- ment in public health benefits (15). Its primary purpose is to ping of CPT tubes (BD Vacutainer) versus shipping whole ensure that basic scientific discoveries translate more rapidly blood, has been investigated. There are data to suggest intonew andbettermedical treatmentsby creatingnew tools that CPT can perform equivalently to Ficoll (19), but it tofindanswersabouthowthesafetyandeffectivenessofnew remains unclear whether shipping spun CPT tubes is medical products can be shown in faster time frames with superior to shipping heparinized whole blood. To allow more certainty and at lower costs. The Critical Path has six for multi-institutional (and multicontinent) trials with areas of focus (including biomarker development and pro- minimal blood sample function loss, the AIDS Network duct manufacturing) and the Critical Areas for Biomarker has established an Immunology Quality Assessment Development are described as follows: Biospecimens, Anal- (IQA) Program to evaluate and enhance the comparabil- ytical Performance, Standardization and Harmonization, ity of immunologic laboratories handling blood samples Bioinformatics, Collaboration and Data Sharing, Stake- from patients (see Appendix 1B and "Centralization of holder Education and Communication, Regulatory Issues, immunologic monitoring," centralized laboratories). and Science Policy. The importance of biomarkers is clear. This approach could be of great benefit to the cancer Good biomarkers offer the prospect for earlier diagnosis, immunotherapy community. focusing expensive and invasive therapies on the right An additional critical issue is the volume of blood populations, monitoring disease progression and therapeu- collected (20). Ideally, banking samples for future ana- tic benefits, as well as facilitating drug development and lyses using newly developed techniques would poten- discovery, clearly many more roles than just surrogate end- tially allow a better understanding of the mechanisms points. Guidelines for incorporation of biomarker studies in of response or exploration of novel prognostic biomar- early clinical trials of novel agents have been published from kers. In a study of 416 blood draws each aimed at taking members of the Biomarkers Taskforce of the NCI Investiga- 250 mL of blood, a median of 200 mL was actually tional Drug Steering Committee (16). The work of the collected in patients with stage III or IV breast cancer. Immunotherapy Biomarkers Taskforce is addressing several The hematocrit of these patients was not significantly challenges specific to immune-based therapies, as follows: decreased during the time of these blood draws, data which may facilitate Institutional Review Board (IRB) 1. Processing and storage of blood samples to bank approvals of larger volume blood draws (20). PBMC peripheral blood mononuclear cells (PBMC) and samples stored for extended periods are being tested serum for immunologic studies. for function (ref. 21; Appendix 1C). 2. Characterization of cellular products for therapy. Recommendations. We recommend following standar- 3. Assay standardization and harmonization before test- dized processing, cryopreservation, storage, and thawing ing patient samples. protocols already tested by the Immunologic Monitoring 4. Centralization of immunologic monitoring. Consortium, or testing the same parameters in your own 5. Standardized (or standardizable) assays which laboratory and stating the extent of standardization in the should be used for clinical trial antitumor immune associated publications. Consider drawing large (200–250 response determination. mL) volumes or carrying out pre- and posttreatment apher- 6. How assay data should be analyzed for "responder" eses, to allow for broad assessment of multiple immune and "nonresponder" identification? parameters including cells and serum and/or plasma. 7. Reporting immunologic monitoring data in publica- Characterization of cellular products for therapy. Chal- tions. lenge. A wide variety of cellular products are being tested 8. Validation of specific assays and/or analytes as bio- for therapy of cancer, from minimally manipulated auto- markers of clinical response. logous blood products to cultured cell lines, and antigen 9. Novel assays in development for immunologic test- loaded, matured dendritic cells (DC; ref. 22). These are ing of patients. required to undergo FDA-mandated tests before release and

3066 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Immunotherapy Biomarkers Recommendations

administration (21 CFR 211.65). Some are relatively different stages involved (28). Assay standardization is straightforward (safety, identity, and purity) and others described in CLIA (Clinical Laboratory Improvements are more complex [potency (developed in phase I and II, Amendments) Guidelines (see Appendix 3; ref. 29) and to be utilized for phase III), stability (acceptable conditions The International Conference on Harmonization of Tech- for both short- and long-term storage), and consistency nical Requirements for Registration of Pharmaceuticals (batch to batch comparability)]. Products that do not meet for Human Use (ICH which brings together the regulatory the prespecified release criteria must not be administered. authorities of Europe, Japan, and the United States and Autologous products can be highly variable between the pharmaceutical industry to discuss scientific and patients and are challenging to characterize and standar- technical aspects of product registration; ref. 30). The dize, and such variability, often minimally characterized, general standardization requirements include quality can impact immune biomarkers. sample processing and storage (see "Processing and Examples. We have included an example of the testing storage of blood samples to bank PBMC and serum for (both exploratory and for product release) conducted for immunologic studies"), a standard operating procedure a current autologous DC-based vaccine clinical trial (in (SOP) for each sample type which maintains sample Appendix 2). The methods for testing safety are well- integrity, and a quality control (QC) program for each standardized. Measures of identity and purity are neces- sample type. SOPs need to be established for each com- sarily specific to the product and are generally flow ponent of the testing process, including sample proces- cytometry based. These might include lineage, activation, sing, the immunoassay, and analysis. For several common and differentiation markers. Potency assays remain cellular immune biomarker assays [cytokine flow cyto- exploratory to date and include testing cell surface and metry, MHC tetramers, and enzyme-linked immunosor- intracellular proteins, cytokines and chemokines pro- bent spot (ELISPOT)], a detailed standard procedure is duced, and activation of target cells [i.e., lymphocyte available (31) and suggestions for standardization are proliferation stimulated by DC vaccines; killing by nat- included. ural killer (NK) cells or T cells]. Assay standardization and validation also involve tech- Two examples of candidate potency assays for antigen- nical performance characteristics of the assay (e.g., repro- presenting cells (APC) are CD54 expression (23) and ducibility intraday and interday, overall precision and interleukin (IL) 12p70 production (24). CD54 upregula- accuracy, specificity, sensitivity, assay range; ref. 16). Cri- tion on the vaccine cells seemed to correlate with overall teria for analytic performance of the assay are as follows: survival in two phase 3 clinical trials (25). Spontaneous the accuracy, precision, reportable range, reference ranges/ and induced IL-12p70 secretionassayshavebeenstan- intervals (normal values), turnaround time, and failure rate dardized (26) and data are now being collected from of the assay as it is to be conducted in the trial. Limits of multiple ongoing DC-based vaccine trials to determine acceptable performance and the QC data should be whether this functional readout correlates with clinical obtained. Data on positive and negative controls, calibra- outcome and could become validated as a potency assay. tors, and reference standards should be available. If the No potency assays have yet been validated and even DC, assay is to be conducted at more than one site, interla- which have been tested in many different clinical trials, boratory variability in the measurements needs to be are sufficiently heterogeneous that each modification in assessed; sources of variation should be minimized to antigen loading and maturation may result in a different maintain performance at all sites within acceptable limits. functional profile; hence, these assays must still be con- Prospective evaluation of assay performance is often exe- sidered exploratory. cuted on a reference population. A bank of PBMCs, serum, Recommendations. Standardize and utilize multiple or leukapheresis products is helpful as a source of material phenotypic and functional assay parameters specific to useful for conducting quality assurance testing (32, 33). the cellular product. In addition to safety, purity, and Commercially available ELISA or Luminex kits are gener- descriptive identity testing for product release, develop- ally well standardized for use. For cellular assays, an exam- ment of candidate potency assays should begin early in ple is characterizing cytomegalovirus (CMV)-specific T-cell clinical testing. Readers are encouraged to refer to FDA’s responses in PBMCs, which allows the identification of low Draft Guidance for Industry (released in October 2008) (<0.5%), intermediate (0.5%–2%), and high (>2%) level Potency Tests for Cellular and Gene Therapy Products responses which could serve as a surrogate for the range of (27). immunity that might be seen in an immunotherapy study Assay standardization and harmonization before testing (34). patient samples. Challenge. Patient blood samples are Establishing clinical utility of the assay involves measur- hard-won resources for understanding the effects of immu- ing the relevant response in immunotherapy trials, that is, notherapy, and are often extremely limited. Therefore, the the magnitude of the association between the assay result primary immune response assays carried out using them and the immunologic and clinical endpoint measured. must be robust and standardized before use in the context Scoring procedures need to be established for quantitative, of a clinical trial. semiquantitative, or qualitative assays (see "Validation of Examples. Preclinical and clinical development phases specific assays and/or analytes as biomarkers of clinical of immunologic biomarkers have been outlined to clarify response").

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3067

Butterfield et al.

In the last several years, harmonization efforts have been Examples. The benefits of these laboratories include organized to understand and reduce variability in ELISPOT, high quality and reliability with QA/QC programs, state- ICS (intracellular cytokine staining), and MHC tetramer of-the-art assay development, standardization, and valida- assays between multiple international laboratories. These tion, as well as decreased cost of immune monitoring (due efforts have involved large proficiency panel programs and to large discounted purchases, batched assay testing, pre- include the option for each laboratory to use their own viously developed assays, and well-trained staff). In addi- materials, reagents, and protocols, to test centrally distrib- tion, assay consultation and result interpretation in uted, pretested PBMCs and antigens with general logistical conjunction with data analysis by biostatisticians can be guidelines to allow for comparison of results obtained from available; and banks of samples for normal controls exist participating labs. These programs allow for the identifica- for comparisons and normal ranges. One drawback is tion of protocol variables which may influence assay timing for sample handling: Shipping from a clinical site outcome. Panel findings have been summarized in harmo- to a central laboratory necessitates a delay of up to 24 hours nization guidelines. An important aspect of these harmo- in sample processing, which may adversely affect PBMC nization efforts is that they do not require strict protocol functional responses as well as the expression of some adherence across all laboratories. Recent findings indicate labile markers on cells and in serum (also ref. 44; e.g., that serum is not required for ex vivo IFN-g ELISPOT, CD62L). Standardized shipping conditions (i.e., materials, according to collaborative studies of different protocols logistics, and monitored temperature) can address some of from the European CIMT Immunoguiding Program and the time-dependent alterations. A different approach is the Cancer Immunotherapy Consortium of the Cancer setting up a regional central laboratory network with cen- Research Institute (33, 35–38), and other international tralized training and oversight (see AIDS network IQA in collaborative studies (32, 38). Appendix 1B). The AIDS Consortium have been leaders in interlabora- The following are 4 examples of multiple centralized tory proficiency testing and in developing programs which laboratories across the United States. The Laboratory of resolve variation between laboratories carrying out immu- Cell Mediated Immunity (LCMI), SAIC-Frederick, Inc., is a noassays (see Appendix 1B; ref. 39). Technician under- centralized contractor laboratory performing immunologic standing of procedures and training has been hurdles to monitoring for the NIH (A.M.). The laboratory is CLIA interlaboratory variation and is surmountable. Use of certified and conducts many assays including modified standard methods and reagents across laboratories also ELISPOT assays including peptide, whole protein, and increases interlaboratory comparability (40). Indeed, stan- tumor cells. Normal donors with known responses to recall dardized training procedures and reagents allowed for antigens serve as positive controls and new assay variations highly concordant IFN-g ELISPOT results to be obtained undergo optimization studies. Cellular Technology Lim- across 7 international laboratories on 3 continents (41). In ited (CTL) operates a GLP-compliant and CLIA-certified another interlaboratory comparison study of the IFN-g laboratory for specimen processing, storage, and immune ELISPOT, 11 assay "novices" were given identical cells, monitoring with an emphasis on ELISPOT. CTL was reagents, and SOPs and the results showed that even with- awarded multiple IDIQ contracts by the NIH for validation out prior experience, standardized procedures and reagents of T-cell–based immune monitoring approaches and has can yield reproducible results across laboratories, coun- served as ELISPOT reference laboratory to the Immune tries, and continents (32). The tuberculosis community Tolerance Network (ITN), and as a central laboratory to (including the WHO Initiative for Vaccine Research) has the Cancer Vaccine Consortium (CVC). The ECOG Central also been involved in ELISPOT assay and related cytokine Immunology Laboratory which operates at the University assay standardization (42, 43). of Pittsburgh Cancer Institute (UPCI; T.L.W. and L.H.B.) Recommendations. We support the use of standardized provides ECOG network investigators with processing, assays (following CLIA and ICH guidelines) with full dis- storage, and testing of specimens under CAP-inspected, closure of methodology specifics for the primary immu- CLIA-compliant conditions. Samples are received via pro- nologic readouts of a trial. Standard protocols and critical tocol-specific specimen shipping kits prepared by the assay parameters for several most commonly used assays laboratory for clinical sites. In E1696, a phase II multi- (particularly for the ELISPOT) have been published (32– epitope peptide vaccine trial for patients with measurable 42) and should be strongly considered for clinical trial metastatic melanoma, a significant difference was reported immune response testing. Participation in external profi- in overall survival of patients stratified by immune ciency panels and use of prescreened PBMCs and other response status, with responders living longer than non- additional controls can be very important for comparing responders (median overall survival ¼ 21.3 vs. 10.8 results between sites and, potentially, between trials. months, P ¼ 0.033; ref. 45). However, in a Cox model Centralization of immunologic monitoring. Challenge. AJCC (American Joint Committee on Cancer) stage at Clinical trial blood and tissue sample processing, cGMP diagnosis was the most significant predictor of overall cellular product production, and immunologic monitoring survival (P ¼ 0.002) and immune response status trended assay standardization are extremely laborious and costly. to significance (P ¼ 0.073). Examples of percent CV (coef- To address these concerns, several institutions have ficient of variation) and assay controls from ECOG 1696 invested in centralized laboratories. ELISPOTs are listed in Appendix 4. Similarly, in an analysis

3068 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Immunotherapy Biomarkers Recommendations

of immune responses to a 12-peptide vaccine conducted in for sufficient amounts of blood, taken at different time the Human Immune Therapy Center at the University of points. Virginia (46), there was a significant correlation between The ELISPOT, perhaps the most thoroughly standar- ELISPOT reactivity and disease-free survival in a univariate dized assay to date (see "Standardized (or standardizable) analysis of 48 patients with resected stages II to IV mela- assayswhichshouldbeusedfor clinical trial antitumor noma who received a multipeptide vaccine in the adjuvant immune response determination"), identifies the number setting. This analysis did not stratify for stage; so other of functional antigen-specific cells, multiple samples may factors could have contributed to this observed correlation. be tested simultaneously, and it can be used for testing Finally, in a trial testing a polyepitope DNA prime/vaccinia more than 1 analyte or function (51). It has been shown boost vaccine in stage III/IV melanoma patients, in which that 2 measures of cytotoxicity,theGranzymeBELISPOT immune monitoring was centrally conducted, a significant Assay (52) and standard 51Cr release correlated better correlation was seen between MHC tetramer responses and with each other than MHC tetramer or IFN-g ELISPOT median survival, and a trend toward correlation between assays, in clinical setting (53). In addition, with auto- IFN-g ELISPOT response and median survival was seen mated analysis methods, great reproducibility and accu- (47). racy for detecting specific T cells can be achieved. An Recommendations. We recommend consideration of alternative functional assay is intracellular staining for central laboratories for immune monitoring, due to their cytokines or other effector molecules using flow cytome- experience in standardized assay conduct and existing try. Multicolor analysis thus provided can complement infrastructure. In particular, they can be a critical part of ELISPOT, as it can provide additional information regard- larger scale, multicenter clinical trials for minimizing var- ing multiple cytokines that are produced by specific sur- iation. It is recommended that central laboratories estab- face-stained cell subsets. Correlation of tetramer assay lish (1) the historical data on any specific standardized results with in vitro cytotoxicity in clinical trial material assay for the selected parameter as the reference and (2) has been previously observed (54). However, in a multi- provide a service to conduct a comparability test to validate center cooperative group trial, MHC tetramer frequencies the data that are generated in other study sites with the and differentiation stage did not correlate with clinical reference data as the control if applicable. outcome but IFN-g ELISPOT response did (Schaefer, Standardized (or standardizable) assays which should be submitted manuscript). used for clinical trial antitumor immune response determina- Another important issue is the target used in the assay. In tion. Challenges. There are many assays potentially cap- two phase II trials of vaccination with a cocktail of altered able of measuring aspects of immune function and limited HLA (human leukocyte antigen)-A2 tumor peptides in blood and tissue samples require choosing a limited num- early melanoma (N ¼ 40) and prostate carcinoma (N ¼ ber of possible assays. The field of immunologic monitoring 20) patients, the ex vivo IFN-g ELISPOT and HLA/peptide is also constantly evolving. The small numbers of complete multimer staining showed a rapid induction of peptide- þ clinical responders, small-scale trials, and variability in specific CD8 T cells of the majority of vaccinated patients. assays chosen and assay conducted make identification of However, clinical efficacy only correlated with significant the crucial assay parameters to measure difficult to identify. antitumor cell activity in vitro. These data clearly stress the Examples. The choice of immunoassay will, in large need for including tumor cell recognition when monitor- part, depend on the proposed mechanisms of action of ing patients treated with tumor vaccines, especially if based the immune intervention. A vaccine designed to generate on modified peptides, to gain better information about a specific antibody response, for example, would focus on tumor reactivity (55, 56). the assessment of humoral immunity rather than a cel- Recent flow cytometric measures of cell-mediated lular response. To test specific immune effectors, there are immunity can evaluate both the target cell death and many choices for the methods such as ELISPOT, MHC- effector cell function simultaneously, allowing for more peptide multimer staining, intracellular cytokine staining, efficient acquisition of both tumor target cell cytolysis as well as soluble cytokines, NK cells, and Th (T helper) and CTL activation. A flow cytometry–based cytotoxicity phenotype. Not only the magnitude of the response and assay has been developed to simultaneously measure NK frequency of effector cells but also the antigenic breadth cell cytotoxicity and NK cell phenotype (57). Another and degree of multifunctionality have all been shown to cytotoxicity assay that has been optimized to utilize low be critical in model systems and specific clinical trials numbers of antigen-specific T cells has been described þ (48) and broad polyepitopic immune responses have (58) in which peptide/MHC multimer-positive CD8 T been associated with complete clinical response in a small cells were purified, cloned, expanded, and tested for group of therapeutically vaccinated patients with vulvar CD107a cell surface expression and their cytotoxicity intraepithelial neoplasia (49). In contrast, vaccine failure evaluated on the basis of the frequency of dead cells in was associated with higher frequencies of disease-specific CMTMR-labeled target cell populations. This assay þ þ CD4 CD25 Foxp3–positive T cells and a low production proved to be more sensitive than the 51Cr-release assay. þ of IFN-g by disease-specific T cells on first vaccination Similarly, combining the measure of CD107a by CD8 (50). The need for several assays to explain the success effector cells with the apoptosis marker Annexin V bind- and failure of this therapeutic vaccine illustrates the need ing to target cells has been used (34, 59, 60).

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3069

Butterfield et al.

There is increasing enthusiasm for polyfunctional flow are unknown and in practice, unknowable—that is, false- cytometric methods, which are already standardized in the positive and false-negative rates are unknown. Use of a HIV community. As discussed at the 2009 workshop, CMV, single standardized definition could allow the results of HIV, and cancer can all induce endogenous T-cell responses different studies to be compared (but will not solve this of varying magnitudes; but only CMV responses tend to be problem). At the UPCI, the Biostatistics Facility (W. Good- protective. The T-cell response signatures for CMV, HIV, ing) developed a definition for individual IFN-g ELISPOT and cancer may be very different: CMV elicits a relatively response that has achieved a measure of success insofar as it þ þ high proportion of IFN-g IL-2 cells with heterogeneous has been shown to correlate with clinical response (45). phenotypes including many effectors. HIV elicits few The definition assumes that there are 3 wells each for þ þ CD8 IL-2 T cells and intermediate phenotypes. Cancer identical test samples (tst) and control samples (ctrl). þ þ patients may show low magnitude, IL-2 but not IFN-g T The variable y is set equal to mean (tst) mean (ctrl). If cells and central memory phenotype. The mechanisms the numbers of cells is different for tst and ctrl, the tst leading to these different signatures need to be further counts are scaled by ratio of the number of tst cells to the elucidated (61). number of ctrl cells, and y is set to 0 if the number of It is essential to develop well-established QC standards responding tst cells in any well is less than the number of that can be made available among different laboratories for responding control cells in any well. An individual’s assay standardization to ensure assay consistency between response to treatment is then determined by the following sites. Minimally, a description of internal (e.g., positive criteria: y (posttreatment) divided by y (pretreatment) antigen or peptide, negative controls, mitogens) and exter- must be greater than 2; and y (posttreatment) must be nal controls need to be provided for each assay (see greater than 10. This definition is essentially based on a "Validation of specific assays and/or analytes as biomarkers factor of 2-fold increase in post- over pretreatment back- of clinical response"). The taskforce members have been in ground-corrected ELISPOT counts, albeit with consider- discussions with the BRB [Biological Resources (Branch)] able protection against false-positive results due to small of the DTP (Developmental Therapeutics Program), NCI, counts. This definition has also been employed in other about creation of a repository for assay standards. This may multicenter trials (64), and an adjustment of it for use be feasible once sources for the agreed-upon standards are with stimulated ELISPOT assays has been shown to have identified. Alternatively, commercial sources are currently low false-positive results, and to correlate with clinical available (SeraCare, Cellular Technology Limited, etc.). outcome(46).Itmustbekeptinmindthatbothtrue Another important aspect of standardization is testing immune response and clinical response are continuous, the extent to which an assay can be run from batched not binary, variables. Different definitions of binary cryopreserved samples, or whether only fresh samples yield response for individual patients are arbitrary and will reliable results. Finally, how to balance standardized assays correlate differently with clinical outcome. Thus, the for immune responses (which allow the field to move relationship of degree of immune response with the forward by having some ability to compare trials) with degree of clinical outcome might be used to refine binary research questions (which drive innovation and may iden- response criteria. Moodie and colleagues (65) have tify novel biomarkers with greater specificity for clinical recently compared response definitions for ELISPOT outcomes)? Larger volumes of blood drawn [without nega- assays. Although they define response in terms of a tive impact on the patient (see "Processing and storage of comparison of test and control samples, the methods blood samples to bank PBMC and serum for immunologic that they describe could generally be adapted to a com- studies")] may allow for this balance with IRB approval. parison of pre- and posttreatment samples. Alternatively, conducting pre- and posttreatment aphereses Recommendations. Obtain multiple pretherapy samples provides cells for monitoring as well as a resource for (at different times) for analysis; these can be used to assess research questions and assay development (62, 63). pretherapy variability of the biomarker level. Tighten Recommendations. We support choosing a standar- response criteria as follows: require positive responses at dized, functional assay as the primary readout of the 2 consecutive posttherapy time points; this is useful for immune response, which addresses the specific hypothesis limiting posttherapy variability. Consider using clinical being tested and proposed mechanism of action of the response to refine the definition of immune response. intervention. There are now strong data that testing multi- When immune response is the primary outcome of interest ple functional parameters (multiple cytokines, recognition in a trial, use nonparametric techniques (such as the of not only peptides but also tumor cells) can yield impor- Wilcoxon signed-ranks test) to assess response of the entire tant information. sample of patents as a group (49). This avoids the problems How assay data should be analyzed for "responder" associated with attempting to define individual responses. and "nonresponder" identification? Challenge. Although To date, no one knows how big an absolute or relative many assays yield data on changes in antitumor immunity, increase in the frequency of antigen-specific T cells between distinguishing assay variation from normal human varia- 2 time points should be considered a biologically relevant tion and treatment-induced responses is not trivial. response (ref. 66 for empirical rules). Examples. All binary response definitions for individual Reporting immunologic monitoring data in publications. patients are effectively "seat of pants;" statistical properties Challenge. Evaluation of immunologic monitoring

3070 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Immunotherapy Biomarkers Recommendations

studies is not possible without disclosure of the details of data (SPICE, M. Roederer, VRC, NIAID, NIH). Peptide the methods that are known to influence test outcome pools are commonly paired as antigen sources for this variability, the assay conduct, and its interpretation. assay and they are evaluated by another shared software Examples. Reporting of complex data sets is a challenge, package (Deconvolute This; ref. 83). as different styles exist that limit comparability. The con- Conventional response criteria may not adequately cept of "minimal information" projects and structured assess the activity of immunotherapeutic agents. Therefore, reporting of data sets was pioneered by Brazma and col- systemic criteria, immune-related response criteria (irRC) leagues (67). The concept has been adapted for T-cell assays have been defined to capture relevant clinical response (68) and multiple minimal information (MI) projects with patterns observed in melanoma patients undergoing overlap now exist. The REMARK criteria have become the immunotherapy. Use of irRC may allow for improved standard for publication of prognostic tumor marker stu- comprehensive evaluation of immunotherapeutic agents dies (69). The REMARK recommendations clearly outline in clinical trials and potentially offer guidance in clinical what should be reported to interpret patient selection, care, as well as being a more appropriate comparator for sample storage, assay performance, and critical statistical correlation with in vitro measures of antitumor immunity analysis. These recommendations are accepted by most (84). Consideration of the time required for evolution of major journals including those published by ASCO and immune responses may require collection of patient blood AACR, yet few immune-based studies adhere to them (69). samples over a longer time period. In the recent FDA draft guidance for therapeutic cancer Recommendations. The immunotherapy field continues vaccines (September 2009), recommendations include to produce novel data from immune assessments in consideration of conducting at least 2 assays and that all patients which correlate to clinical outcome in different assay parameters and controls should be clearly described diseases and treatment settings. These candidate biomar- (70–78). It is recommended that investigators make them- kers should first be standardized (85) and then validated by selves familiar with these different projects to choose the other investigators. The evolution of antitumor immunity appropriate guidelines for the assays chosen. may necessitate longer term immunologic monitoring. Recommendations. We suggest that the following study Novel assays in development for immunologic testing of aspects should always be included when reporting results, patients. Challenge. To move toward more sensitive, independent of other applicable guidelines: The QA/QC high-throughput evaluations, there must first be quality conducted, reference populations included, all reagents sample acquisition for analysis and hypothesis testing. and controls tested, at least some selected examples of Also, it is not only immunotherapy trials which must be truly representative raw data and the assay performance evaluated. Most biological agents used singly or in combi- characteristics. These parameters will allow appropriate nation with conventional drugs for cancer therapy engage reviewer and reader evaluation of the quality and potential the immune system. Other biological agents that specifi- impact of the data. cally target growth factor receptors, blood vessels/endothe- Validation of specific assays and/or analytes as biomarkers lial cells, tumor cells, or tumor-associated antigens often of clinical response. Challenge. When data from basic involve immunologic mechanisms. and translational research settings (and exploratory test- Examples. Directly assaying the tumor environment, ing in phase I trials) suggest that a specific immunoassay performing expression arrays (from the tumor), testing or other biomarker correlates with clinical outcome, for determinant/epitope spreading, and testing genetic standardization and validation are needed to substantiate aspects of the host [single nucleotide polymorphism those data and to allow possible comparisons between (SNP), genome-wide association studies (GWAS), HLA] treatments and trials. are not yet commonly conducted. One example of a Examples. GLP guidelines from the FDA for general large-scale initiative to analyze patient tumors for indivi- laboratory conduct for assay performance are available dual gene expression patterns is "M2Gen," a research (79). For the ELISPOT assay, validation of the assay has collaboration between H. Lee Moffitt Cancer Center and been addressed (80). A recent report (81) has described Merck & Co. Researchers are collecting tumor tissues from many aspects of standardization of MHC tetramer, IFN-g patients to identify the biological markers unique to each ELISPOT, and IFN-g real-time PCR and the authors also tumor (86). attempted to validate the assays for determining the abso- Immune profiling can include the following: high- lute frequency of antigen-specific T cells. They concluded throughput molecular profiling platforms to study the that this could not be accomplished without a "gold human immune system, polychromatic flow cytometry, standard" measure for such cells. RNA profiling (mRNA, miRNA, RNAseq), SNP arrays (soon Quantification of polyfunctional cytokine-producing T genome sequence), multiplex serum chemokines, cyto- cells (often IFN-g, IL-2, and TNF) by multicolor flow kines profiles, and protein and peptide arrays (87) for cytometry is being used in infectious disease models and serologic responses, mapping antigenic repertoire, and HIV patients for correlation with clinical outcome (60, 82). semiquantitative immunohistochemistry of the tumor This is an immunologic readout which may also serve as a (88). In addition to the positive effects of antitumor effec- biomarker for clinical response and there is already stan- tor activation, critical aspects of tumor immunosuppres- dardized shared software for analysis of the flow cytometric sion should be investigated as the frequency and function

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3071

Butterfield et al.

of MDSC (myeloid-derived suppressor cells) and Treg between human subjects’ protection committees and (regulatory T cells), functional defects in TIL and in circu- patient advocate representatives along with researchers lating immune cells, cytokine imbalance (Th2 vs. Th1), and clinicians. failure to generate central memory T cells, persistent activa- As a service to the community, in addition to the present tion of T cells, spontaneous apoptosis of T cells, T-cell report, we propose the following steps: senescence, and presence of soluble factors in serum that a. Specific SOP recommendations (many of which have induce death in immune cells. All of these immune deficits been standardized and some of which are published) have been reported but are not yet part of the regular with links to be posted on the iSBTc-SITC web site for immune monitoring repertoire. This is crucial for under- easy access (and also to other society web sites). standing why some approaches are not successful and for b. Promoting greater focus on standardization and vali- personalized selection of available anticancer therapies in dation assay guidelines [CLSI (Clinical and Labora- the future. tory Standards Institute), CLIA, ICH]. An important aspect of these broad assessments of c. Improved data reporting immune biomarker studies immunity, particularly with newer, high-throughput should adhere to published guidelines, especially approaches, is data management (89). Currently data those that are advocated by the journals to which are often stored in multiple clinical and laboratory data- the papers are submitted such as REMARK. Many bases requiring manual data entry and coordination. efforts are ongoing to refine these recommendations Informatics must address the following: the ability to MIFlowCyt, MIACA, MIATA, MIBBI, that will help to integrate data from multiple technology platforms, the select the "MI" projects most applicable to a specific ability to integrate clinical and biomarker data from scientific question or setting. multiple projects, and include an emphasis on data dis- d. Further discussion at the follow-up workshop: "Sym- semination/high availability (to allow for downstream posium on Immuno-Oncology Biomarkers, 2010 analyses by biostatistics/bioinformatics teams; for access/ and Beyond: Perspectives from the iSBTc-SITC Bio- query by investigators, and ultimately to promote insight, marker Task Force," Masur Auditorium on the NIH sharing data with study participants, collaborators, con- campus, Building 10, Clinical Center, Bethesda, MD, sortia members, scientific community, and streamline September 30, 2010 (90). data export to public repositories). A goal inherent to e. Strengthening of interimmunotherapy society com- this is defining a universal data element set to accompany munication and collaborations (Disis, Fox, manu- all high-quality biospecimens. script in preparation). Recommendations. We recommend that both RNA and Although not directly in the hands of individual inves- DNA samples as well as sera and plasma be banked under tigators and smaller teams, we also recommend greater standardized conditions for later testing in multiplex, funding levels to support the acquisition of blood and molecular assays (from blood and the tumor and to study tumor samples for embedded correlative studies as well as the microenvironment). Improved collection of tumor and TIL are crucial for understanding the impact of different unspecified banking for future analysis. It is only with therapeutic approaches. We also reiterate that sufficient resources of tumor, serum/plasma, PBMCs, DNA, and blood be drawn to allow for the planned testing of the RNA that we will be able to learn as much as possible primary hypothesis being investigated in the trial, such that about the state of immunity in cancer patients, the positive additional baseline and posttreatment blood is banked for effects of our interventions, and the inhibitory effects of testing novel hypotheses (or generating new hypotheses) tumor progression that we have yet to overcome. These freshly tested and banked samples, collected and assayed that arise in the field during the time required for trial design, approvals, enrollment, and conclusion. under standardized conditions, will also be crucial in allowing us to better understand patient-to-patient varia- Discussion bility and take steps toward more effective and personalized approaches.

Immunotherapy clinical trials can only benefit from Appendix 1A careful study of the effects on patient immune responses and the state of immune function (and dysfunction). Sample Collection, Cryopreservation, and Thawing Because of the large variation between patients, elimina- (Lymphocytes/PBMCs) tion of as much variation as possible in procedures used Factors that Did Not Matter. for handling blood and tumor specimens, and in procedures for assays, is essential. Equally essential is the * Shipping on dry ice, 24, 48, 72 hours versus liquid thorough reporting of the level of standardization and nitrogen (P > 0.05). the specific methods used for specimens, assays, and * Large (100 mL) versus small volume (25 mL) dilution analysis. We are also recommending an increased level for thaw (P > 0.05). of banking of diverse biological specimens for unspeci- * Time (5–10 minutes) and speed of spin at wash, 1,200 fied future research. We recognize that implementation to 1,500 rpm (P > 0.05). of this goal will require discussion and cooperation * Number of cells per vial (1, 2, 3 10e7; P > 0.05).

3072 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Immunotherapy Biomarkers Recommendations

Factors that Did Matter. The thawing method. Addi- Viability. The cells are counted by microscopic obser- tives such as human serum albumin, dextran, and FBS were vation on a hemacytometer and a differential count (DCs superior to human AB serum; washing thawed cells in vs. lymphocytes) is obtained using trypan blue dye. Mini- medium prewarmed to 25Cto37C was superior to mum 70% viability. chilled (4C) medium (91). Purity. The DCs must express MHC class II and CD86 Open Access Protocols are available from the Immuno- by flow cytometry in a minimum of 70% of the cells. logic Monitoring Consortium (91, 92). Additional phenotyping (MHC class I, CD80, CD83, CCR7, others) is conducted to fully characterize the DC Appendix 1B and is for research proposes. Sterility. DCs are tested by bacterial (aerobic and anae- robic) and fungal cultures at the Clinical Microbiology Examples from the AIDS Network Laboratory. Final results of the microbial cultures are IQA Program. The IQA is a resource designed to help available in 14 days. Prior to release of the DC for vaccine immunologists evaluate and enhance the integrity and use, a standard Gram stain is conducted and must be comparability of immunologic laboratory determinations negative for the presence of microorganisms. conducted on patients enrolled in multisite HIV/AIDS Mycoplasma testing of cell suspensions (not superna- investigations. tants) is conducted using a rapid detection system, based on nucleic acid hybridization or by PCR. The cell prepara- * About 83 participating laboratories. tion must be negative for mycoplasma. * 6 test shipments per year. Endotoxin testing is conducted on the cell culture at the * The viability of PBMCs and the viable yield before time of harvest and prior to release of the final product. The freezing and after thawing is tested. acceptable endotoxin level is less than 5 EU/kg of body * IQA reviews statistical report and identifies sites that weight per dose. are having difficulty performing assays. Potency. To define a measure of potency for the DC, we * Poor performers are contacted to discuss specific pro- determine their ability to produce IL-12p70 and IL-10 by blems. Luminex assay (26). This test is conducted batched, with * Laboratories fax histogram results to the IQA for and without activation by CD40L and/or LPS, and is central review. available several weeks after vaccine injection. Data will Laboratories who serve both the Aids Clinical Trials be correlated with measures of DC phenotype and clinical Group (ACTG) and IMPAACT trials must conform to the outcome. stringent criteria of the ACTG (80% viability and 80% In addition, a 0.5-mL sample of the final DC preparation viable cell recovery of PBMCs; refs. 93, 94). from each vaccination time is cryopreserved for possible ancillary testing in the future. These samples are stored a Appendix 1C minimum of 1 year after vaccine administration. Large volume blood samples from cancer patients (median volume collected: 200 mL; median yield: 0.8 10e6 Appendix 3 PBMCs/mL; ref. 21). Clinical Laboratory Improvements Amendments (29) Validation in stored samples: ongoing performance Test accuracy (close agreement to the true value), Analysis of 80 samples Precision (agreement of independent results: same day, ¼ Sample age range: 30–2,500 days, average 600 different day), Median recovery: 70% Reproducibility (intra-assay and interassay) Mean recovery: 70% Reportable range (limits of detection) Range: 22%–130% Normal ranges (pools of healthy donors, accumulated patient samples: test at least 20, include a banked healthy donor control in patient assays), Personnel competency testing (minimally, annually) Appendix 2 Equipment validation, monitoring Reagent tracking Cellular Product Minimal Safety, Purity, and Identity Tests The following is an example of the specific release tests Appendix 4 which are required by the FDA for early phase trials invol- ving autologous, in vitro manipulated cellular products (in Preliminary examples from the ECOG Central Immu- this example, DC). This example also shows the identity/ nology Laboratory ELISPOT assays of large-scale, multi- purity testing chosen for this type of product, and the institution clinical trials [n ¼ 20 from each trial, from >70 candidate potency test being conducted. (E1696) and >200 (E4697) patient ELISPOT assays]

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3073

Butterfield et al.

E4697 (2008–2009) Spontaneous PMA/I ()/OKT3 Center for Biologics Evaluation and Research) and the FDA liaison to iSBTc-SITC for providing critical critique of the manuscript. They also thank the staff of iSBTc-SITC for production and technical assistance in Healthy control 4.9 (54% CV) 304 (19.2% CV the manuscript preparation: Tara Withington,CAE,ExecutiveDirector; average intra-assay) Angela Kilbert, Director of Administration; Chloe Surinak, Project Man- ager; Roseann Marotz, Meetings Manager; and Jimmy Balwit, Scientific (48% CV interassay) Communications. Patient average 0.7 (35% CV) 81 (38.7% CV) – E1696 (2002 2003) Spontaneous PMA/I /PHA Grant Support

Healthy control 5.4 (56% CV) 284 (15.5% CV The study was supported by NIH grants P50 CA121973 and RO1 CA104524 to average intra-assay) L.H. Butterfield; NIH grants RO1 CA119123 and R21 CA123864 to B.A. Fox; and research grant ("Standardization of Immune Monitoring") to C.M. Britten and S. (51% CV interassay) H. van der Burg from the Wallace Coulter Foundation, Miami, FL. This project has Patient average 19 (40% CV) 171 (18.8% CV) been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, Disclosure of Potential Conflicts of Interest commercial products, or organizations imply endorsement by the U.S. Govern- ment. The iSBTc-SITC/FDA/NCI Workshop on Immunotherapy Biomarkers was S. Janetzki is founder and president of ZellNet Consulting, Inc., a for- supported by educational grants obtained by iSBTc-SITC from Bristol-Myers profit company offering ELISPOT plate evaluation and consulting services, Squibb Company, the Center for Cancer Research, National Cancer Institute, and is the Coordinator of the Immune Assay Working Group of the Cancer Merck Oncology, Biogen Idec, the Center for Human Immunology, Autoimmu- Immunotherapy Consortium, Cancer Research Institute, a non-profit orga- nity, and Inflammation, and with additional support from Pfizer and Immuneer- nization. The other authors disclosed no potential conflicts of interest. ing Corporation.

Acknowledgments Received August 19, 2010; revised November 12, 2010; accepted The authors thank Dr. Raj Puri (Director, Division of Cellular and December 10, 2010; published OnlineFirst May 10, 2011. Gene Therapies, Office of Cellular, Tissue and Gene Therapies, FDA/

References 1. Dendreon. www.dendreon.com. Available from: http://investor.den- mulatory blockade in dendritic cell-based melanoma immunotherapy. dreon.com/phoenix.zhtml?c = 120739&p = irol-newsArticle&ID = J Immunother 2004;27:354–67. 1398598&highlight¼. 12. Keilholz U, Weber J, Finke JH, Gabrilovich DI, Kast WM, Disis ML, 2. Kantoff PW HC, Shore ND, Berger ER, Small EJ, Penson DF, Redfern et al. Immunologic monitoring of cancer vaccine therapy: results of a CH, et al. Sepuleucel-T immunotherapy for castration-resistant pros- workshop sponsored by the Society for Biological Therapy. J Immun- tate cancer. N Engl J Med 2010;363:411–81. other 2002;25:97–138. 3. Available from: http://www.asco.org/ASCOv2/Meetings/Abstracts? 13. Available from: http://www.cancer.gov/images/trwg/immune_oct08. &vmview¼abst_detail_view&confID¼65&abstractID¼32035. pdf. 4. Schwartzentruber DJ, Lawson D, Richards J, Conry RM, Miller D, 14. Available from: http://www.fda.gov/downloads/Drugs/Guidance- Triesman J, et al. A phase III multi-institutional randomized study of ComplianceRegulatoryInformation/Guidances/ucm079855.pdf. immunization with the gp100;209–217 (210M) peptide followed by 15. Available from: http://www.fda.gov/ScienceResearch/SpecialTopics/ high-dose IL-2 compared with high-dose IL-2 alone in patients with CriticalPathInitiative/default.htm. metastatic melanoma. J Clin Oncol 2009;27:18s. 16. Dancey JE, Dobbin KK, Groshen S, Jessup JM, Hruszkewycz AH, 5. Schuster SJ, Neelapu SS, Gause BL, Muggia FM, Gockerman JP, Koehler M, et al. Guidelines for the development and incorporation of Sotomayor EM, et al. Idiotype vaccine therapy (BiovaxID) in follicular biomarker studies in early clinical trials of novel agents. Clin Cancer lymphoma in first complete remission: phase III clinical trial results. Res 2010;16:1745–55. ASCO Meeting Abstract. J Clin Oncol 2009;27:2. 17. Bull M, Lee D, Stucky J, Chiu YL, Rubin A, Horton H, et al. Defining 6. Morgan RA, Dudley ME, Wunderlich JR, Hughes MS, Yang JC, Sherry blood processing parameters for optimal detection of cryopreserved RM, et al. Cancer regression in patients after transfer of genetically antigen-specific responses for HIV vaccine trials. J Immunol Methods engineered lymphocytes. Science 2006;314:126–9. 2007;322:57–69. 7. Rosenberg SA, Restifo NP, Yang JC, Morgan RA, Dudley ME. Adop- 18. Kierstead LS, Dubey S, Meyer B, Tobery TW, Mogg R, Fernandez VR, tive cell transfer: a clinical path to effective cancer immunotherapy. et al. Enhanced rates and magnitude of immune responses detected Nat Rev Cancer 2008;8:299–308. against an HIV vaccine: effect of using an optimized process for 8. Lee KH, Wang E, Nielsen MB, Wunderlich J, Migueles S, Connors M, isolating PBMC. AIDS Res Hum Retroviruses 2007;23:86–92. et al. Increased vaccine-specific T cell frequency after peptide-based 19. Ruitenberg JJ, Mulder CB, Maino VC, Landay AL, Ghanekar SA. vaccination correlates with increased susceptibility to in vitro stimula- VACUTAINER CPT and Ficoll density gradient separation perform tion but does not lead to tumor regression. J Immunol 1999;163:6292– equivalently in maintaining the quality and function of PBMC from HIV 300. seropositive blood samples. BMC Immunol 2006;7:11. 9. Butterfield LH, Comin-Anduix B, Vujanovic L, Lee Y, Dissette VB, Yang 20. Available from: http://www.irb.pitt.edu/irbmailings/listbot6-20-05. JQ, et al. Adenovirus MART-1-engineered autologous dendritic cell htm. vaccine for metastatic melanoma. J Immunother 2008;31:294–309. 21. Disis ML, Dang Y, Slota MS, Childs J, Higgins D, Bates N, et al., and 10. Butterfield LH, Ribas A, Dissette VB, Amarnani SN, Vu HT, Oseguera Members of the Immunologic Monitoring Consortium. [cited 2009] D, et al. Determinant spreading associated with clinical response in Available from: http://www.isbtc.org/meetings/access.php?content¼ dendritic cell-based immunotherapy for malignant melanoma. Clin workshop09pres&loc¼/meetings/am09/workshop09/presentations/ Cancer Res 2003;9:998–1008. Disis_915am.pdf. 11. Ribas A, Glaspy JA, Lee Y, Dissette VB, Seja E, Vu HT, et al. Role of 22. Nicolette CA, Healey D, Tcherepanova I, Whelton P, Monesmith T, dendritic cell phenotype, determinant spreading, and negative costi- Coombs L, et al. Dendritic cells for active immunotherapy: optimizing

3074 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Immunotherapy Biomarkers Recommendations

design and manufacture in order to develop commercially and clini- human immunodeficiency virus vaccine clinical trials by peripheral cally viable products. Vaccine 2007;25 Suppl 2:B47–60. blood mononuclear cell and enzyme-linked immunospot assays in 23. Sheikh NA, Jones LA. CD54 is a surrogate marker of antigen pre- laboratories from three continents. Clin Vaccine Immunol 2009; senting cell activation. Cancer Immunol Immunother 2008;57:1381– 16:147–55. 90. 42. Smith SG, Joosten SA, Verscheure V, Pathan AA, McShane H, 24. Mailliard RB, Wankowicz-Kalinska A, Cai Q, Wesa A, Hilkens CM, Ottenhoff TH, et al. Identification of major factors influencing ELI- Kapsenberg ML, et al. alpha-type-1 polarized dendritic cells: a novel Spot-based monitoring of cellular responses to antigens from Myco- immunization tool with optimized CTL-inducing activity. Cancer Res bacterium tuberculosis. PLoS One 2009;4:e7972. 2004;64:5934–7. 43. Hanekom WA, Dockrell HM, Ottenhoff TH, Doherty TM, Fletcher H, 25. Higano CS, Schellhammer PF, Small EJ, Burch PA, Nemunaitis J, McShane H, et al. Immunological outcomes of new tuberculosis Yuh L, et al. Integrated data from 2 randomized, double-blind, vaccine trials: WHO panel recommendations. PLoS Med 2008;5:e145. placebo-controlled, phase 3 trials of active cellular immunotherapy 44. Avache S, Panelli M, Marincola FM, Stroncek DF. Effects of storage with sipuleucel-T in advanced prostate cancer. Cancer 2009;115: time and exogenous protease inhibitors on plasma protein levels. Am 3670–9. J Clin Pathol 2006;126:174–84. 26. Butterfield LH, Gooding W, Whiteside TL. Development of a potency 45. Kirkwood JM, Lee S, Moschos SJ, Albertini MR, Michalak JC, Sander assay for human dendritic cells: IL-12p70 production. J Immunother C, et al. Immunogenicity and antitumor effects of vaccination with 2008;31:89–100. peptide vaccineþ/-granulocyte-monocyte colony-stimulating factor 27. Draft guidance for industry-potency tests for cellular and gene therapy and/or IFN-alpha2b in advanced metastatic melanoma: Eastern products. [cited 2008]. Available from: http://www.fda.gov/cber/ Cooperative Oncology Group Phase II Trial E1696. Clin Cancer Res gdlns/testcellgene.htm. 2009;15:1443–51. 28. Dang Y, Disis ML. Identification of immunologic biomarkers asso- 46. Slingluff CL Jr, Petroni GR, Chianese-Bullock KA, Smolkin ME, Hib- ciated with clinical response after immune-based therapy for cancer. bitts S, Murphy C, et al. Immunologic and clinical outcomes of a Ann N Y Acad Sci 2009;1174:81–7. randomized phase II trial of two multipeptide vaccines for melanoma 29. Available from: http://www.cms.hhs.gov/CLIA/05_CLIA_Brochures. in the adjuvant setting. Clin Cancer Res 2007;13:6386–95. asp. 47. Dangoor A, Lorigan P, Keilholz U, Schadendorf D, Harris A, Ottens- 30. ICH. Available from: http://www.ich.org/LOB/media/MEDIA417.pdf. meier C, et al. Clinical and immunological responses in metastatic 31. NCCLS. Performance of Single Cell Immune Response Assays: Pro- melanoma patients vaccinated with a high-dose poly-epitope vac- posed Guidelines. NCCLS document I/LA26-P [ISBN 1-56238-507-0]. cine. Cancer Immunol Immunother 2010;59:863–73. Vol. 23, Number 25. Wayne, PA: NCCLS.; 2003. 48. Welters MJ, Kenter GG, Piersma SJ, Vloon AP, Lowik€ MJ, Berends- 32. Zhang W, Caspell R, Karulin AY, Ahmad M, Haicheur N, Abdelsalam A, van der Meer DM, et al. Induction of tumor-specific CD4þ and CD8þ et al. ELISPOT assays provide reproducible results among different T-cell immunity in cervical cancer patients by a human papillomavirus laboratories for T-cell immune monitoring–even in hands of ELISPOT- type 16 E6 and E7 long peptides vaccine. Clin Cancer Res inexperienced investigators. J Immunotoxicol 2009;6:227–34. 2008;14:178–87. 33. Janetzki S, Panageas KS, Ben-Porat L, Boyer J, Britten CM, Clay TM, 49. Kenter GG, Welters MJ, Valentijn AR,LowikMJ,Berends-vander et al. Results and harmonization guidelines from two large-scale inter- Meer DM, Vloon AP, et al. Vaccination against HPV-16 oncoproteins national Elispot proficiency panels conducted by the Cancer Vaccine for vulvar intraepithelial neoplasia. N Engl J Med 2009;361:1838–47. Consortium (CVC/SVI). Cancer Immunol Immunother 2008;57:303–15. 50. Welters MJ, Kenter GG, de Vos van Steenwijk PJ, Lowik€ MJ, Berends- 34. Maecker HT, Hassler J, Payne JK, Summers A, Comatas K, Gha- van der Meer DM, Essahsah F, et al. Success or failure of vaccination nayem M, et al. Precision and linearity targets for validation of an for HPV16-positive vulvar lesions correlates with kinetics and phe- IFNgamma ELISPOT, cytokine flow cytometry, and tetramer assay notype of induced T-cell responses. Proc Natl Acad Sci U S A using CMV peptides. BMC Immunol 2008;9:9. 2010;107:11895–9. 35. Britten CM, Gouttefangeas C, Welters MJ, Pawelec G, Koch S, 51. Quast S, Zhang W, Shive C, Kovalovski D, Ott PA, Herzog BA, et al. IL- Ottensmeier C, et al. The CIMT-monitoring panel: a two-step 2 absorption affects IFN-gamma and IL-5, but not IL-4 producing approach to harmonize the enumeration of antigen-specific CD8þ memory T cells in double color cytokine ELISPOT assays. Cell T lymphocytes by structural and functional assays. Cancer Immunol Immunol 2005;237:28–36. Immunother 2008;57:289–302. 52. Rininsland FH, Helms T, Asaad RJ, Boehm BO, Tary-Lehmann M. 36. Britten CM, Janetzki S, Ben-Porat L, Clay TM, Kalos M, Maecker H, Granzyme B ELISPOT assay for ex vivo measurements of T cell et al. Harmonization guidelines for HLA-peptide multimer assays immunity. J Immunol Methods 2000;240:143–55. derived from results of a large scale international proficiency panel 53. Shafer-Weaver K, Rosenberg S, Strobl S, Gregory Alvord W, Baseler of the Cancer Vaccine Consortium. Cancer Immunol Immunother M, Malyguine A. Application of the granzyme B ELISPOT assay for 2009;58:1701–13. monitoring cancer vaccine trials. J Immunother 2006;29:328–35. 37. Britten CM, Janetzki S, Van Der Burg SH, Gouttefangeas C, Hoos A. 54. Dhodapkar MV, Steinman RM, Sapp M, Desai H, Fossella C, Kra- Toward the harmonization of immune monitoring in clinical trials: quo sovsky J, et al. Rapid generation of broad T-cell immunity in humans vadis? Cancer Immunol Immunother 2008;57:285–8. after a single injection of mature dendritic cells. J Clin Invest 38. Mander A, Gouttefangeas C, Ottensmeier C, Welters MJ, Low L, van 1999;104:173–80. der Burg SH, et al. Serum is not required for ex vivo IFN-gamma 55. Iero M, Filipazzi P, Castelli C, Belli F, Valdagni R, Parmiani G, et al. ELISPOT: a collaborative study of different protocols from the Eur- Modified peptides in anti-cancer vaccines: are we eventually improv- opean CIMT Immunoguiding Program. Cancer Immunol Immunother ing anti-tumour immunity? Cancer Immunol Immunother 2009;58: 2010;59:619–27. 1159–67. 39. Fahey JL, Aziz N, Spritzler J, Plaeger S, Nishanian P, Lathey JL, et al. 56. Malyguine A, Strobl SL, Shafer-Weaver KA, Ulderich T, Troke A, Need for an external proficiency testing program for cytokines, che- Baseler M, et al. A modified human ELISPOT assay to detect specific mokines, and plasma markers of immune activation. Clin Diagn Lab responses to primary tumor cell targets. J Transl Med 2004;2:9. Immunol 2000;7:540–8. 57. Kim GG, Donnenberg VS, Donnenberg AD, Gooding W, Whiteside TL. 40. Denny TN, Gelman R, Bergeron M, Landay A, Lam L, Louzao R, et al. A A novel multiparametric flow cytometry-based cytotoxicity assay North American multilaboratory study of CD4 counts using flow simultaneously immunophenotypes effector cells: comparisons to a cytometric panLeukogating (PLG): a NIAID-DAIDS Immunology Qual- 4 h 51Cr-release assay. J Immunol Methods 2007;325:51–66. ity Assessment Program Study. Cytometry B Clin Cytom 2008; 58. Devevre E, Romero P, Mahnke YD. LiveCount Assay: concomitant 74 Suppl 1:S52–64. measurement of cytolytic activity and phenotypic characterisation of 41. Boaz MJ, Hayes P, Tarragona T, Seamons L, Cooper A, Birungi J, CD8(þ) T-cells by flow cytometry. J Immunol Methods 2006;311: et al. Concordant proficiency in measurement of T-cell immunity in 31–46.

www.aacrjournals.org Clin Cancer Res; 17(10) May 15, 2011 3075

Butterfield et al.

59. Zaritskaya L, Shafer-Weaver KA, Gregory MK, Strobl SL, Baseler M, 77. Taylor CF, Field D, Sansone SA, Aerts J, Apweiler R, Ashburner M, Malyguine A. Application of a flow cytometric cytotoxicity assay for et al. Promoting coherent minimum reporting guidelines for biological monitoring cancer vaccine trials. J Immunother 2009;32:186–94. and biomedical investigations: the MIBBI project. Nat Biotechnol 60. Maecker HT, Dunn HS, Suni MA, Khatamzas E, Pitcher CJ, Bunde T, 2008;26:889–96. et al. Use of overlapping peptide mixtures as antigens for cytokine 78. [cited 2009]. Available from: www.fda.gov/biologicsbloodvaccines/ flow cytometry. J Immunol Methods 2001;255:27–40. guidancecomplianceregulatoryinformation/default.htm. 61. Nomura L, Maino VC, Maecker HT. Standardization and optimization 79. Available from: http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/ of multiparameter intracellular cytokine staining. Cytometry A cfCFR/CFRSearch.cfm. 2008;73:984–91. 80. Cox JH, Ferrari G, Janetzki S. Measurement of cytokine release at 62. Dols A, Smith JW II, Meijer SL, Fox BA, Hu HM, Walker E, et al. the single cell level using the ELISPOT assay. Methods 2006;38: Vaccination of women with metastatic breast cancer, using a costi- 274–82. mulatory gene (CD80)-modified, HLA-A2-matched, allogeneic, breast 81. Xu Y, Theobald V, Sung C, DePalma K, Atwater L, Seiger K, et al. cancer cell line: clinical and immunological results. Hum Gene Ther Validation of a HLA-A2 tetramer flow cytometric method, IFNgamma 2003;14:1117–23. real time RT-PCR, and IFNgamma ELISPOT for detection of immu- 63. Walker EB, Haley D, Petrausch U, Floyd K, Miller W, Sanjuan N, et al. nologic response to gp100 and MelanA/MART-1 in melanoma Phenotype and functional characterization of long-term gp100-spe- patients. J Transl Med 2008;6:61. cific memory CD8þ T cells in disease-free melanoma patients before 82. Koup RA, Roederer M, Lamoreaux L, Fischer J, Novik L, Nason MC, and after boosting immunization. Clin Cancer Res 2008;14:5270–83. et al. Priming immunization with DNA augments immunogenicity of 64. Slingluff CL Jr, Petroni GR, Olson WC, Smolkin ME, Ross MI, Haas recombinant adenoviral vectors for both HIV-1 specific antibody and NB, et al. Effect of GM-CSF on circulating CD8þ and CD4þ T cell T-cell responses. PLoS One 2010;5:e9015. responses to a multipeptide melanoma vaccine: outcome of a multi- 83. Roederer M, Koup RA. Optimized determination of T cell epitope center randomized trial. Clin Cancer Res 2009;15:7036–44. responses. J Immunol Methods 2003;274:221–8. 65. Moodie Z, Price L, Gouttefangeas C, Mander A, Janetzki S, Lower€ M, 84. Wolchok JD, Hoos A, O'Day S, Weber JS, Hamid O, Lebbe C, et al. et al. Response definition criteria for ELISPOT assays revisited. Guidelines for the evaluation of immune therapy activity in solid Cancer Immunol Immunother 2010;59:1489–501. tumors: immune-related response criteria. Clin Cancer Res 2009; 66. Dubey S, Clair J, Fu TM, Guan L, Long R, Mogg R, et al. Detection of HIV 15:7412–20. vaccine-induced cell-mediated immunity in HIV-seronegative clinical 85. Kalos M. An integrative paradigm to impart quality to correlative trial participants using an optimized and validated enzyme-linked science. J Transl Med 2010;8:26. immunospot assay. J Acquir Immune Defic Syndr 2007;45:20–7. 86. Available from: http://www.moffitt.org/Site.aspx?spid¼ 67. Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, C54AF116F69244D49BACE202F69BC2A6. Stoeckert C, et al. Minimum information about a microarray experi- 87. Gaseitsiwe S, Valentini D, Mahdavifar S, Magalhaes I, Hoft DF, ment (MIAME)-toward standards for microarray data. Nat Genet Zerweck J, et al. Pattern recognition in pulmonary tuberculosis 2001;29:365–71. defined by high content peptide microarray chip analysis repre- 68. Available from: http://www.miataproject.org. senting 61 proteins from M. tuberculosis. PLoS One 2008;3: 69. McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. e3840. Reporting recommendations for tumor marker prognostic studies 88. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce- (REMARK). J Natl Cancer Inst 2005;97:1180–4. Pages C, et al. Type, density, and location of immune cells within 70. Alvarez DF, Helm K, Degregori J, Roederer M, Majka S. Publishing human colorectal tumors predict clinical outcome. Science 2006;313: flow cytometry data. Am J Physiol Lung Cell Mol Physiol 2010;298: 1960–4. L127–30. 89. Chaussabel D, Ueno H, Banchereau J, Quinn C. Data management: it 71. Hoos A, Parmiani G, Hege K, Sznol M, Loibner H, Eggermont A, et al. A starts at the bench. Nat Immunol 2009;10:1225–7. clinical development paradigm for cancer vaccines and related bio- 90. Butterfield LH, Disis ML, Khleif SN, Balwit JM, Marincola F. Immuno- logics. J Immunother 2007;30:1–15. oncology biomarkers 2010 and beyond: Perspectives from the iSBTc/ 72. Available from: http://mibbi.org/index.php/MIBBI_portal. SITC Biomarker Task Force. J Transl Med 2010;8:130. 73. Available from: http://www.mibbi.org. 91. Disis ML, dela Rosa C, Goodell V, Kuan LY, Chang JC, Kuus- 74. Janetzki S, Britten CM, Kalos M, Levitsky HI, Maecker HT, Melief CJ, Reichel K, et al. Maximizing the retention of antigen specific lym- et al. "MIATA"-minimal information about T cell assays. Immunity phocyte function after cryopreservation. J Immunol Methods 2006; 2009;31:527–8. 308:13–8. 75. Lee JA, Spidlen J, Boyce K, Cai J, Crosbie N, Dalphin M, et al. 92. Maecker HT, Moon J, Bhatia S, Ghanekar SA, Maino VC, Payne JK, MIFlowCyt: the minimum information about a Flow Cytometry Experi- et al. Impact of cryopreservation on tetramer, cytokine flow cytometry, ment. Cytometry A 2008;73:926–30. and ELISPOT. BMC Immunol 2005;6:17. 76. Spidlen J, Moore W, Parks D, Goldberg M, Bray C, Bierre P, et al. Data 93. Available from: http://aactg.org/about-actg. File Standard for Flow Cytometry, version FCS 3.1. Cytometry A 94. Available from: http://iqa.center.duke.edu/modules/dhvi_iqa/index. 2010;77:97–100. php?id¼1.

3076 Clin Cancer Res; 17(10) May 15, 2011 Clinical Cancer Research

Recommendations from the iSBTc-SITC/FDA/NCI Workshop on Immunotherapy Biomarkers

Lisa H. Butterfield, A. Karolina Palucka, Cedrik M. Britten, et al.

Clin Cancer Res 2011;17:3064-3076. Published OnlineFirst May 10, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-10-2234

Cited articles This article cites 75 articles, 15 of which you can access for free at: http://clincancerres.aacrjournals.org/content/17/10/3064.full#ref-list-1

Citing articles This article has been cited by 13 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/17/10/3064.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/17/10/3064. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.