Evaluation of Dendritic Cell Products Generated for Human Therapy and Post-Treatment Immune Monitoring

ABSTRACT

Immunotherapy with autologous dendritic cells (DCs) engineered ex vivo to become immunogenic or tolerogenic recently has been used in an array of human diseases. However, optimizing the DC generation process to achieve a product with maximal clinical efficacy has been challenging. Manufacturing issues to be resolved include the type and characteristics of DC lots generated in patients with different diseases; the type of antigens or cells used for DC loading; the combination of cytokines and media used for DC maturation; and assay selection for measuring DC potency. Standardized release criteria for clinical-grade DC are not yet available, and given the current state of knowledge about DC biology, may be premature to mandate. Similarly, to evaluate immunological efficacy following adoptive transfers of antigen-loaded DC, standardized monitoring assays are needed. Although several single-cell–based immune assays and multiplex cytokine platforms are available, the choice and quality control of monitoring assays remain problematic. Improvements in the manufacturing and evaluation of clinical-grade DC products and in the design and quality of monitoring assays are needed to adequately assess the therapeutic impact of DC-based interventions in future clinical trials.

Recent advances in molecular immunology and preclinical studies in experimental animals suggest that immunizations based on transfers of antigen-loaded dendritic cells (DCs) offer promise for therapy of malignant diseases. In fact, DCs have been used in active immunotherapy of cancer for more than a decade, and DC-based products have been delivered to more than 1,000 cancer patients enrolled in more than 150 clinical trials.^1–3 Active immunotherapy with DCs for human diseases, including cancer, is based on the accumulated evidence that DCs represent nature's best antigen-presenting cell (APC). DCs recognize, process, and present foreign antigens to T cells in the effector arm of the immune system.⁴ They can drive primary and secondary immune responses. Immature DCs capture and internalize antigens via receptor and nonreceptor-mediated mechanisms.⁵ A complex molecular array of cell components referred to as the antigen-processing machinery (APM) processes the internalized antigens to yield peptides that are subsequently expressed on the cell surface in association with major histocompatibility complex (MHC) molecules and β₂ -microglobulin as a trimolecular complex.⁶ Mature DCs, equipped with surface molecules facilitating cellular interactions, present these peptides to cognate T cells.^7–9 When the antigenic peptides are tumor derived, activation of tumor-specific cytolytic T lymphocytes (CTLs) occurs, and the immune system is primed to recognize the tumor and possibly to eliminate it. Clearly, the processing of antigens by DCs and their presentation to T cells are key to successful immunizations in cancer and other diseases.

A confocal microscope image of human immature dendritic cell (DC) stained with labeled antibodies specific for CD11c (red) and HLA-DR (green). The nucleus (blue) is stained with DAPI.

Subtypes of Dendritic Cells

The division of labor evident among DCs suggests that the DC population is heterogenous. DCs can be broadly divided into two subtypes—myeloid and plasmacytoid—based on phenotype and function.^10,11 Myeloid DCs are monocyte-derived DCs that are functionally categorized as DC1. They polarize T cells toward T helper (Th1) functions and can mediate both MHC I– and MHC II–restricted presentation of tumor epitopes, which is considered necessary for optimal immunotherapy of cancer. ^3,12–14 Plasmacytoid DCs have DC2 activity and polarize T cells toward Th2 functions. Because DC-driven polarization determines the quality of T-cell responses, the source and type of DCs selected for active immunotherapy are very important. Because plasmacytoid DCs tend to induce tolerance, they are considered to be an appropriate cell type for treating autoimmune diseases. Thus, the DC type best suited for active immunotherapy may vary depending on therapeutic application.

Clinical Results to Date

DC products have been used for therapy of infectious diseases, including chronic HIV-1 infections,^15,16 autoimmune diseases,¹⁷ post-transplant graft versus host disease (GVHD),¹⁸ and various cancer types.^13,14 Results from numerous Phase 1 trials indicate that DC-based immunotherapy is feasible, safe, and well tolerated. However, clinical results are often inconclusive, in part because the subjects enrolled in these trials had an advanced-stage disease but also because of the lack of product standardization, resulting in tremendous phenotypic and functional differences in administered therapeutic DC products.

Abbreviations

Obtaining DCs

DCs can be obtained in large numbers from in vitro cell culture of either peripheral blood monocytes or bone marrow progenitors. The most common method for obtaining clinical-grade DCs in vitro from leukapheresis product uses CD14+ monocytes (or plastic adherent monocytes) and growth factors that include granulocyte macrophage colony stimulating factor (GM-CSF) and interleukin (IL)-4 or GM-CSF plus IL-13.^7,9 The other approach includes the use of CD34+ cells with combinations of stem-cell factor, GM-CSF, Flt3-ligand, plus tumor necrosis factor-alpha (TNF)-α.

Production Improvements

Improvements in the design and manufacture of DC products for therapy are being actively addressed. Efforts to optimize DC-based therapy for clinical use include the establishment of quality control measures for consistent generation of high-quality therapeutic vaccines as well as for reliable monitoring of responses to DC-based vaccines in patients receiving this therapy. The major objective of this review is to discuss strategies for the evaluation of therapeutic DC products at the time they are ready for adoptive transfer and for the development of methods that would reliably measure in vivo effectiveness of DC in inducing desirable immune responses in subjects receiving DC-based therapy.

DC GENERATION

Figure 1 presents a schema for generating human DC from peripheral blood monocytes. Apheresis products are processed by elutriation using an Elutra cell separation system (Gambro BCT, Lakewood, CO) to recover a monocyte fraction. The fraction purity is ascertained using flow cytometry by determining the frequency of CD14+ cells. Culture of the recovered monocytes in the presence of IL-4 (1,000 IU/mL) and GM-CSF (1,000 IU/mL) for six days yields immature DCs (iDCs). These cells excel in antigen uptake or endocytosis.⁹ Complex antigens taken up by iDCs are processed by APM in the cytosol. Next, iDCs are matured in the mix of cytokines (IL-1b, IL-6, TNF-α, and interferons) to yield mature DCs (mDCs). The latter are excellent peptide-presenting cells and represent a final product that can be released for therapy, provided it meets established release criteria.

Figure 1. A schema for in vitro generation of human dendritic cells (DCs) from peripheral blood monocytes

The method of DC generation described above may vary, depending on how the monocytes are isolated from peripheral blood mononuclear cells and on conditions adopted for their culture and maturation. Monocyte purity, culture medium, cytokine content, and culture vessels exert considerable influence on the quality of the final DC product.²¹ Production methods vary from manual approaches using plastic flasks to semi-automated functionally closed systems for monocyte culture. The latter offer a simple and economical method for meeting good manufacturing practice (GMP) requirements and facilitate process standardization and the reproducible production of high-quality DCs.

LOADING ANTIGENS ONTO DCs

A DC product designed for therapy must be loaded with tumor antigens to complete the manufacturing process and to generate APCs that elicit the desirable anti-tumor responses in vivo after their adoptive transfer. A variety of defined and undefined antigens have been used for DC loading (Table 1). Defined antigens, such as peptides, proteins, cDNA, or mRNA, may elicit tumor-specific T-cell responses targeting a well-defined epitope overexpressed in vivo in tumor cells. Defined antigens also allow for targeted immune monitoring following vaccination.

Table 1. Tumor-derived antigens commonly used for DC loading in preparing vaccines

Undefined antigens, including apoptotic tumor cells, tumor cell lysates or extracts, tumor cell–DC fusions, and total genomic DNA or total mRNA derived from tumor cells, have the potential to generate anti-tumor responses targeting a broad spectrum of epitopes in the tumor.

Both defined and undefined antigens have disadvantages (Table 1). Therefore, selecting a DC payload is not simple, and requires an understanding of the antigenic characteristics of the tumor as well as the immunogenic potential of the tumor antigens under consideration. It is essential to remember that complex antigens have to be processed by DCs before presentation, and should be delivered to iDCs, wherease mDCs are preferred for pulsing with peptides (Table 1). Technical issues relevant to the most efficient delivery of proteins, DNA, or mRNA to DC also must be considered.

CHALLENGES IN GENERATING DC PRODUCTS FOR THERAPY

Regardless of the DC source and culture methods used, therapeutic-grade DCs should meet standard criteria for release. However, no universal criteria for DC release exist, and each laboratory is obliged to define its own release standards based on product sterility, viability, purity, and stability. In the absence of defined universal standards, the quality of DC products is likely to vary between production facilities, and a considerable number of variables must be addressed.^22,23 For example, in addition to the source of DC precursor cells, which may be autologous or allogeneic, the type of tumor antigen(s) used for DC loading, the DC maturation process, their activation state and potency, as well as their ex vivo stability, are among the key variables in the development of DC-based products.

To ensure that a safe and effective DC product will be available for therapy, it is necessary to implement and adhere to SOPs based on experience from preclinical studies and the cumulative current knowledge of DC biology. Endotoxin-free materials are essential for DC production, because even transient exposure of DC precursor monocytes to endotoxin impairs subsequent IL-12 production,²⁴ which seems to be a mandatory factor for effective DC function. Given the variables that are likely to influence the quality of the final product, a central manufacturing site is needed to ensure consistency and quality in processing and in the final product. A facility with HEPA-filtered biological safety cabinets, incubators, centrifuges, cryopreservation devices, and freezers is necessary. The use of components and materials meeting GMP criteria is also required as clinical studies progress from Phase 1 to Phase 3. This type of facility is most likely to meet the challenge of producing a high-quality, optimally functional DC product.

In addition to the optimized production process, monitoring DC potency before adoptive transfer is essential for establishing clinically relevant endpoints. Immune monitoring of patients enrolled in DC-based vaccination protocols represents a special challenge that requires expertise in the performance of serial-type assays. Immune monitoring before, during, and after DC-based vaccine delivery is a critical component of the process required for evaluating the immunologic and clinical effects of the DC-based vaccines.

ESTABLISHING THE QUALITY OF THERAPEUTIC DC PRODUCTS

The quality of therapeutic DC products has to be assessed before they can be released for clinical use. This is usually accomplished by: first determining the phenotypic characteristics of DCs using multicolor flow cytometry that confirms cell maturity and expression of co-stimulatory and migration-related molecules on DC; and then measuring the DCs' ability to induce antigen-specific CTL from naive peripheral blood mononuclear cells (PBMC) precursors. The latter is difficult to execute because of the requirement for functional T-cell assays (e.g., cytotoxicity or ELISPOT), and for prior IVS with DCs as well as the availability of defined antigens or targets. For this reason, various surrogate methods for measuring DC functions have been introduced, including intercellular adhesion molecule (ICAM) expression, cytokine production, endocytosis of latex beads or mannose receptor expression on the cell surface.^25,26 These methods are useful in defining the functional status of DC, but do not provide information relevant to their ability to mediate antigen-specific T-cell responses.

Most current anti-tumor vaccines depend on the use of ex vivo manipulated autologous products, thus representing "personalized" therapy. A DC product generated from cells of one individual is referred to as a DC batch. Patient-to-patient variability in starting material makes it difficult to ensure DC batch uniformity. Inter-batch uniformity would be easier to achieve with normal allogeneic DCs than with autologous DC products. Although "off-the-shelf" allogeneic DC products might be more practical and uniform, the possibility of viral transmission and rapid in vivo clearance of allogeneic cells introduces another set of concerns and regulatory issues. Once a batch of DCs is made and its quality tested, it can be aliquoted into sterile vials, and the cells cryopreserved for multiple vaccinations. However, any further manipulation of the vaccine, such as washing or adjusting cell numbers, will necessitate testing for sterility and endotoxin levels before vaccine delivery. From a practical viewpoint, the same batch of DCs can deliver multiple vaccines provided no additional manipulations of DCs are performed.

Current Standards for Release of Therapeutic Dendritic Cell Products

Although there are no widely accepted release criteria for therapeutic DCs, the cells should meet predefined sterility, viability, purity, and stability standards (according to 21 CFR 211) as well as defined phenotypic and functional criteria. Assays used for DC characterization and product release are summarized in Table 2. Current FDA recommendations for release of therapeutic DC products specify sterility, viability, purity, phenotypic characterization, and stability measurements, but no functional assays for Phase 1 and 2 studies (FDA current GTP final rule; 21 CFR 211). A definition of release criteria based on expected results of these assays is made before the implementation of DC production and is adhered to for all DC batches produced by a facility. For example, if DC viability of 80% is defined as a release standard, any DC batch with viability <80% would be rejected. The development of potency assays for DC is encouraged throughout the early-phase clinical trials, but is not a requirement for DC release.

Table 2. Procedures and assays required to evaluate dendtric cell products generated for therapy

Sterility testing. Sterility testing of DC products requires 14-day incubation and ensures the absence of aerobic or anaerobic microorganisms. Results are not available until after DC administration. Therefore, a Gram stain is performed immediately before DC release for therapy. Negative results from the Gram stain and the availability of the established plan of action necessary in case a contaminated DC product is administered are sufficient criteria for release of DC products. The products also must be mycoplasma free. In this instance, DNA hybridization, or RT-PCR–based assays, which take several hours to complete, are performed instead of 28-day mycoplasma culture assays. Endotoxin levels are measured in the final DC product, which must be endotoxin free. For the mycoplasma and endotoxin assays, medium and cells must be submitted for testing.

Viability, purity, and stability. Viability, measured microscopically by a trypan blue-dye test or flow cytometry, using PI or 7 amino-actinomycin D (7-AAD), is necessary to show that most DCs in culture are alive. Viability measurements are especially useful after freezing and thawing DCs aliquoted for multiple vaccinations. Purity determinations use flow cytometry to evaluate the presence of lymphocytes (CD3+, CD19+ or CD56/16+) and/or monocytes (CD14+) in a DC product. In general, purity of 80% or higher is desirable, although the overall requirement for the purity level still remains to be established and will be dictated by the overall immunopotency of the final product. Stability assays measure DCs' ability to retain viability and activity over time at room temperature or in the cold, as well as possible detrimental effects of freezing and thawing. Stability assays are especially critical when a DC-based product is divided into smaller batches that are cryopreserved to be thawed for administration to patients for multiple dosing over long treatment intervals.

Functional assays. A variety of functional assays for newly generated DC products are available (Table 3). One popular assay is the mixed lymphocyte reaction (MLR), in which allogeneic T cells serve as responders and DCs—titered into the co-culture at various T cell–DC ratios—serve as stimulators.²⁹ The assay measures levels of T-cell responses to alloantigens presented by DCs, and is not directly relevant to the response elicited by tumor antigens. A requirement for the panel of T cells that can recognize a variety of HLA specificities represents a considerable practical limitation for a routine use of this assay. Therefore, flow cytometry-based assays have largely replaced MLR for functional assessments of DCs. Among these, endocytosis of FITC-labeled dextran 30 combined with measurement of mannose receptor expression with anti-CD206 antibody is often used.²⁵ This assay measures an entirely different function of DCs, one that is essential for antigen uptake and that potentially discriminates between a receptor-mediated antigen internalization versus endocytosis. These two pathways of antigen uptake may have different consequences for antigen processing and presentation.³¹ Therefore, it might be important to determine their use by tumor-derived antigens relative to the type of T-cell response these antigens elicit in vivo.

Table 3. Assays for functional evaluation of dendritic cell products*

Flow cytometry assays for DCs. Flow cytometry-based assays measure the state of DC maturation. After five to six days of culture, immature DCs are CD40+, CD80+, CD86+, and HLA-DR+, but have low expression levels of CD83. Upon maturation, DCs typically express higher levels of HLA molecules and co-stimulatory molecules (CD80, CD83, and CD86), in addition to enhanced expression of the chemokine receptor, CCR7. The presence of CD206 (the mannose receptor) and absence of CD14 (a monocyte marker) on the DC surface may be helpful for defining the state of DC differentiation. Of the utmost importance is full ex vivo activation of DCs, which can be achieved by various approaches (e.g., combinations of cytokines, toll-receptor ligands, or the CD40 ligand) and can be evaluated by upregulation of surface markers and by functional assays.

More recently, methods able to quantitate expression levels of Class I and Class II major histocompatibility complex (MHC) molecules, ICAM, or individual components of APM have been introduced to assess DCs' potential for T-cell stimulation.³² These flow-based assays proved to be quite informative, as DCs derived from monocytes of patients with cancer were shown to have aberrant expression of APM components, which was related to inadequate T-cell stimulation, at least in vitro. ³² The ability to evaluate DC-mediated Th1/Th2 polarization (e.g., expression of interferon (IFN)-gamma or IL-4) following co-culture of purified CD3+CD45RA+ lymphocytes with DC and re-stimulation with PMA/ionomycin also has been used in some studies.³³

Measuring co-stimulatory capacity. In addition to processing antigens and presenting individual epitopes to epitope-specific T cells, DCs exhibit a strong co-stimulatory capacity, which is much higher than that of other cells. In fact, expression of co-stimulatory molecules on DC has been used as a surrogate marker for their functional potential.³⁴ Because co-stimulation plays a critical role in the induction of tumor-specific immunity, this assay may be relevant to the ability of DCs to function in stimulating T cells specific for tumor antigens. As such, the assay for co-stimulatory molecule expression, called the costimulatory (COSTIM) assay, might be especially useful in assessing the functional potential of ex vivo-generated DCs in anticancer vaccines. It has been reported to work with both allogeneic and autologous DC products.³⁵

General remarks on DC release criteria. While the various functional assays listed in Table 3 provide useful information about DC properties, none has been validated as a tool for measuring the potency of DCs generated for human therapy. These assays might be standardized, but a formal validation would require establishing a correlation with the in vivo performance of the cells. It appears that none of the listed assays has been correlated with the clinical efficacy of the administered DC products. Nevertheless, the availability of several assays that measure diverse DC functions provides an opportunity for an improved control of the variability of DC products and for establishing a clinically meaningful functional profile for DCs. A functional DC profile based on results of these assays is likely to give a more reliable estimate of DCs' therapeutic potential than any single assay and may prove to be a better surrogate of their clinical efficacy.

DEVELOPMENT OF A POTENCY ASSAY FOR HUMAN DENDRITIC CELLS

Although DCs are known to secrete multiple cytokines, the ability to produce IL-12p70 spontaneously or upon activation with CD40L, and with or without addition of innate immunity signals (e.g., LPS) is considered essential for optimal in vivo activity. Importantly, the data from animal models of tumor growth indicate that IL-12p70 is the cytokine responsible for DC-mediated Th1-polarization and for the enhancement of T lymphocyte anti-tumor responses (Table 4).^27,28 The ability of DCs to produce and secrete IL-12p70 has been used to develop and standardize a DC potency assay.²⁶

Table 4. Rationale for selecting IL-12 production assay to measure potency of dendritic cells

A potency assay can be defined as a quantitative measure of DCs' biological function(s) assessed in vitro and in vivo. The potency assay for IL-12p70 production, which recently has been introduced and standardized for routine use in the author's laboratory, consists of two steps (Figure 2). First responder DCs are co-incubated with J588 lymphoma cells stably transfected with the human CD40 ligand gene as stimulators.²⁶ Second, supernatants are tested from these co-cultures in the Luminex system to determine levels of IL-12p70 secreted by DCs stimulated with J558/CD40L with or without LPS. Both steps of this potency assay required standardization, and considerable effort has been invested in defining the assay's performance. This assay has an interassay coefficent of variation of 18.5% (n = 30), and a broad dynamic range, which facilitates evaluation of various DC products characterized by vastly different levels of IL-12p70 production. The normal range for the assay established using DC products generated from peripheral blood monocytes of 13 normal donors and matured in a conventional cytokine cocktail (IL-1b, IL-6, TNF-α, PGE2) was from 8 to 999 pg/mL, with a mean of 270 pg/mL.²¹ DCs generated from patients with melanoma or a chronic HIV-1 infection and matured in the same cytokines produced vastly different levels of IL-12p70 (range: 11 to 777 pg/mL and 22 to 1,605 pg/mL, respectively). Not surprisingly, culture conditions and maturation cytokines used for DC production were the major determining factors in the IL-12p70 production level. Thus, DC matured in the presence of a cytokine cocktail containing α- or γ-interferons and polyI:C produced high levels of IL-12p70 in subjects with brain cancer (range: 8 to 11,600 pg/mL) or those with Sezary's syndrome (range: 630 to 15,900 pg/mL). The in vivo therapeutic and immunologic efficacy of DC products that contain cells secreting different levels of the cytokine is currently unknown and is being evaluated in immunotherapeutic trials to assess the biologic significance of IL-12 secretion levels. The objective is to establish a correlation between IL-12p70 production levels by DCs used for therapy and clinical endpoints. Ultimately, this will provide in vivo validation of the IL-12p70 assay. It is expected, but not yet proven, that DC products able to produce high IL-12p70 levels following ex vivo stimulation will mediate superior priming or antigen-presenting functions, resulting in significant clinical responses.

Figure 2. Diagrammatic representation of a two-step potency assay measuring IL-12p70 production by DC. The assay description appears in the text.

ASSESSMENTS OF ANTIGEN-SPECIFIC DENDRITIC CELL FUNCTIONS

Once DCs are loaded with antigen, several additional methods are available for evaluating antigen-specific immune responses in vitro (Table 3).

Single-cell Assays

Assays that measure DCs' ability to induce activation and expansion of antigen-specific T cells are of particular value. Among them, single-cell T-cell assays,³⁶ which determine the frequency of T cells capable of responding to an epitope presented by DCs by cytokine production (e.g., ELISPOT assay or CFC) or MHC-tetramer staining are often used to evaluate DC products. Correlations between the results of these assays and clinical outcomes have been difficult to demonstrate, however, and there are no standardized approaches across laboratories for performing these assays. Nevertheless, these single-cell assays could provide the necessary objective data for optimizing DC-based immunotherapy. ELISPOT assays with antigen-specific T-cell lines as responder cells can evaluate DCs' ability to present the relevant antigen in vitro, in which a defined tumor antigen or peptide represents the immunogen. However, it is often not possible to perform such assays in a clinical setting, as antigen-specific or tumor-specific T cells usually are not available.

Population-based Assays

Several population-based assays measuring either proliferation of autologous T cells co-incubated with antigen-loaded DCs or target-cell cytotoxicity, in which CTLs generated in co-cultures (IVS) are tested for the ability to lyse the relevant target, are available for DC product characterization. These population-based assays are generally less informative than single-cell assays. These assays can be used not only for assessments of the DC-based products' potential to ex vivo stimulate T-cell responses, but also for immune monitoring of PBMC obtained from cancer patients treated with DC products.

In the absence of a defined antigen test, the capability of DCs to present antigens to autologous T cells can be assessed using recall antigens, such as CMV or influenza virus antigens, that re-stimulate memory T-cell responses. Pre-existing immunity against one or both of these antigens is widespread. Therefore, DCs obtained from individuals immunized with these antigens could be loaded with the antigens, allowing for the re-stimulation of the corresponding autologous T cells in vitro and confirmation of DC function.³⁷ However, these tests do not confirm the ability of the DCs to invoke primary immune responses, which may be a critical requirement for DCs to achieve high immune potency.

SELECTING ASSAYS TO MONITOR DENDRITIC CELL–BASED CANCER VACCINES

Immune monitoring of vaccination protocols requires assays that can accurately measure vaccine-induced changes in the frequency and function of antitumor effector T cells. Assays selected for monitoring of immune cells following therapy must be adaptable to serial testing with a minimal loss of accuracy. It is important to select assays that can accurately discriminate between therapy-induced changes in an immune response relative to that measured at baseline.

Many new types of immune assays are available for immune monitoring of DC-based clinical trials. It is possible to prioritize phenotypic versus functional, specific versus non-specific, direct versus indirect, and single cell versus bulk assays (Table 5).³⁶ The choice of assays for immune monitoring often depends on available resources. Current emphasis has been on targeted assays that can evaluate specific activation pathways or even individual signaling molecules. Assays that identify and measure the frequency of subsets of cells or individual cells engaged in a response to a specific stimulus are of special interest. In general, tumor antigen-specific assays now can be performed reliably and precisely are replacing older and less informative nonspecific tests. Multiplex assays that define an immunologic profile (e.g., Th1 versus Th2 cytokine profile), are more desirable than assays able to measure only one analyte. A better understanding of immune mechanisms and hypothesis-driven monitoring are being combined in establishing immune measures as biomarkers of tumor progression or patient survival. The objective is to reliably use immune measures as surrogate endpoints of clinical responses to biotherapy. It is, however, unrealistic to expect that a single assay will recapitulate in vivo responses to the DC-based vaccine because several mechanisms contribute to vaccine-related upregulation of immune responses following DC transfers. Some of these mechanisms might be orchestrated by the tumor, for example, and they could suppress anti-tumor immune activity.³⁸ Thus, the knowledge of mechanisms operating in the tumor microenvironment is an important component of immune-monitoring strategies.

Table 5. Assays currently available for immune monitoring of vaccine-based protocols

Once an assay that accurately reflects vaccine-induced changes in the phenotype or function of immune cells is selected, it may require adaptation to serial monitoring may be to evaluate cryopreserved "batched" specimens. Specimen batching is required to test all serial samples collected from a patient in the course of therapy in the same assay. This practice avoids inter-assay variability and increases reliability. This also requires the laboratory to compare the assay performance on both fresh and cryopreserved/thawed cells before its acceptance for monitoring. Certain assays, notably those that measure cytotoxicity, cannot be performed reliably with cryopreserved/thawed mono-nuclear cells.³⁹ Assays that must be performed on freshly harvested specimens require documentation of inter-assay variability so that vaccine-induced changes in an immune measure can be distinguished from assay-related daily variability. Reliable assays performed with cryopreserved/thawed specimens are the best candidates for serial monitoring of DC-based cancer vaccines.

THE DTH TEST

To date, no in vivo measures of immune response to the DC-based vaccines are available, except for delayed-type hypersensitivity (DTH) skin tests, which are not commonly used. Reluctance to incorporate the DTH skin test into vaccine monitoring is related to the necessity for measurements of skin induration, which must be performed 48–72 hours after antigen application, and requires the patient's return to the clinic. It is possible to measure DTH to recall antigens (PPD, tetanus, candida) or to specific epitopes used in a vaccine, or both. As such, the DTH skin test represents a useful strategy for assessing the general status of recall immunity and for determining the likelihood of mounting a response to the vaccine. It has been reported that a good qualitative relationship exists between DTH responses and ex vivo T-cell responses to peptides used for vaccination in patients with cancer.⁴⁰ Most important, a conversion from a negative to positive skin test to the vaccine is an excellent sign of vaccine-induced immunity, and it has been reported to correlate with clinical responses in patients with melanoma. For these reasons, consideration should be given to a more frequent use of DTH tests measuring responses to vaccine components and administered before and after vaccination. A positive DTH skin test to the vaccine after its administration provides a strong rationale for immune monitoring, using peripheral blood or tumor-draining lymph nodes, if the latter are available. Previously, we have described the strategy depending on DTH skin test results for selecting monitoring assays.⁴¹

Use of DTH to Evaluate Anti-tumor Immunity

When designing immune monitoring for DC-based vaccination trials, it is necessary to measure vaccine-specific responses, although responses to PMA/ionomycin, anti-CD3 antibody, or mitogens measured in parallel could be used as a general guide for immune responsiveness of an individual patient. There are two reasonable initial approaches: measure antibody responses to the vaccine, and look for a shift from IgM to IgG responses in serial serum specimens; or, measure proliferative responses of lymphocytes (or cytokine production) to the vaccine in bulk cultures. Both can be considered as first-line screening assays. Pre-vaccine and post-vaccine specimens should be tested together in the same assays.

If both assays give negative results, but a DTH skin test is positive, further ex vivo testing is required, because it may be possible that the bulk assays are not sensitive enough to detect antigen- or peptide-specific responses. The decision must be made whether to proceed with single-cell assays such as ELISPOT, cytokine flow cytomery (CFC), or tetramer assay(s) without in vitro sensitization (IVS) or to resort to single-cell assays performed after IVS. This is a difficult decision because if the frequency of antigen- or peptide-specific T cells is low, then direct (no IVS) single-cell assays could be negative. On the other hand, the IVS strategy allows for in vitro expansion of antigen- or peptide-specific T cells before the assay, but it is time-consuming, labor-intensive, and costly.

If both screening assays give positive results, consider single-cell assays without IVS for determining the frequency of antigen- or peptide-responsive T cells. A lack of systemic in vivo response to antigens (peptides) used in the vaccine suggests that ex vivo assays also will be negative and that the patient is unlikely to respond immunologically to the vaccine. Nevertheless, immune monitoring with the most sensitive methods is necessary to formally document the absence of cellular immunologic responses. In such circumstances, the availability of a standardized, robust, single-cell assay that measures function and cellular phenotype is important.

SINGLE-CELL ASSAYS IN MONITORING DC-BASED VACCINATION TRIALS

Single-cell assays that detect the frequency of epitope-specific effector T lymphocytes have been used to evaluate immune responses to DC-based cancer vaccines. Among these, CFC, tetramer binding, and ELISPOT measure the frequency of these lymphocytes in the mononuclear cell specimens obtained from the peripheral circulation of vaccinated individuals.^42–44 All three assays are based on TCR recognition of cognate peptides presented by MHC Class I or Class II molecules on the surface of DC-presenting antigens to the responder T cells. No consensus exists, however, as to which of the three assays should be used to monitor vaccination results. The perception that these assays are equivalent (i.e., that they provide the same results) may not be correct. The authors have compared the performance of the assays in monitoring the frequency of melanoma peptide-specific CD8+ T cells in the peripheral circulation of subjects with metastatic melanoma who had received multi-epitope DC-based vaccines.⁴⁵ Concordance among the three assays was estimated using a 3 x 3 scatter plot matrix design constructed for each of the four peptides tested in all three assays before and after vaccination therapy was completed. The three single-cell assays were not found to be concordant in measuring the frequency of immune effector cells in the peripheral blood of vaccinated subjects. The results for tetramer staining were consistently higher than those obtained with the ELISPOT or CFC assays,⁴⁰ indicating either that tetramers bound to peptide-specific T cells unable to secrete cytokines or that they bound to nonspecific peptide T cells. Similar observations have been reported by others.⁴⁶

Advantages and Disadvantages of Single-cell Assays

The ELISPOT assay measures the production of cytokines (most commonly IFN-γ or IL-5) by individual T cells in the plated population, with a theoretical detection sensitivity of 1/100,000 cells.⁴⁷ CFC identifies single-responding T cells (1/50,000) with expression of a cytokine in the Golgi zone. Tetramer binding detects peptide-specific T cells expressing the relevant TCR with a theoretical detection sensitivity of 1/10,000 cells. The assays not only have different sensitivities of detection, but also differ in specificity. ELISPOT and CFC are antibody-based and are highly specific. In contrast, tetramers, which are complexes of peptides sitting in grooves of four MHC molecules held together by a streptavidin-biotin scaffold,⁴⁸ bind to T lymphocytes expressing the relevant TCR with variable affinity. Tetramers might easily dissociate or non-specifically bind to B cells, monocytes, or apoptotic cells.^49,50 Tetramer specificity needs to be carefully controlled. T cells that bind tetramers may not be functional, as TCR signaling could be compromised, as often happens in cancer.^46,51 This reduces the tetramer-binding assay to a phenotypic category because the assay detects T cells that bind tetramers but may not be functional.⁴⁶ CFC measures cytokine expression in a cell and not its secretion (although it is commonly assumed that the expressed cytokine would be secreted). Cell permeabilization necessary for intercellular staining of a cytokine in the Golgi zone might introduce problems with immunodetection in CFC assays.

ELISPOT is based on a similar principle as CFC, but it measures cytokine secretion from stimulated responder cells that are plated as a monolayer of individual cells on a nitrocellulose membrane to avoid cell-to-cell contact and allow for adequate spot display. ELISPOT measures the function of individual responder cells by identifying those that produce and secrete the measured cytokine. ELISPOT does not require cell permeabilization or the use of a flow cytometer for cytokine detection.⁴³ Measuring function rather than phenotype is preferable, making ELISPOT an assay of choice.

However, CFC and tetramer binding are flow cytometry-based assays and allow for surface labeling of responder cells and their identification. It is possible to select CD8+ or CD4+ T-cell subsets on antibody-charged columns before ELISPOT,⁵² and two-color ELISPOT now available offers the possibility of identifying T cells simultaneously producing two cytokines.⁵³ In addition, supernatants from ELISPOT wells can be collected and tested for cytokine levels in multiplex assays. On the other hand, tetramer binding can be combined with surface staining to determine cellular phenotype and intracytoplasmic staining to detect cytokine production.⁴⁶ While very informative, especially in situations when some tetramer-binding cells do not express cytokines, this technology is time-consuming and labor-intensive and thus not the best choice for serial monitoring. The recommended solution would be to monitor by ELISPOT or CFC (but not both), depending on sample numbers, time, labor, cost, and access to a flow cytometer, and then use tetramer binding as a confirmatory assay in situations where it is important to demonstrate a functional deficiency of tetramer-binding T cells. ELISPOT performed under strictly controlled, standardized conditions provides accurate estimates of the frequency of functionally competent effector T cells in batched, serial samples obtained from subjects enrolled in clinical studies with DC-based vaccines. Compared to CFC and tetramer binding, the cost of ELISPOT permits its use in a high-volume testing. However, the ELISPOT assay is not easy to standardize, and responder-stimulator interactions might result in unacceptably high background spot counts, making the assay uninterpretable.

CONCLUSIONS

Dendritic cell (DC) products for cancer patient therapy are being used in clinical trials to enhance anti-tumor immune responses, which are often compromised in patients with advanced disease. DC therapies are also being used in patients with post-transplant graft versus host disease (GVHD) or patients with infectious diseases.^15,16,18 Studies in animal models of autoimmunity suggest that in the future, adoptive transfers of tolerogenic DCs may be beneficial to patients with autoimmunity.¹⁷ DC-based therapies are nontoxic, but their clinical efficacy has not been confirmed.⁵⁴ In part, this result might reflect the complexity of DC production processes and the lack of universal criteria for quality and release of therapeutic DC products. Currently, manufactured DC products are phenotypically and functionally variable, and products made in different laboratories are not comparable. Although this might be due to the source and functional competence of DC precursors in cancer patients, differences in production methods clearly contribute to DC product variability. To standardize DC production, several manufacturing issues must be addressed, including the selection of media and conditions for DC culture, the composition of maturation cytokines, and the choice and standardization of assays used.

As more information is being generated about the biologic characteristics of DCs, significant changes in the manufacturing process are being made. At the same time, DC products generated for therapy must meet predefined criteria for sterility, viability, purity, and stability. These products also must be defined phenotypically and functionally, and strict attention should be paid to their activation and maturation status, as those characteristics likely correlate with clinical endpoints.

Reliable production of biologically active, clinically effective DC products with defined potency will require modifications of current methods, including the use of semi-automatic culture devices for clinical-scale production, the addition of novel cytokines, and the introduction of improved antigen uptake methods. Equally important to the future success of DC-based therapies is monitoring of patients' immune responses after vaccination. Immune monitoring of DC-based cancer vaccines is essential for establishing correlations between clinical endpoints and tumor-specific as well as vaccine-specific immunity. Although elusive, such correlations have been increasingly frequently reported in recent cancer vaccination trials,^55,56 perhaps because of more sophisticated vaccine designs or improved quality of immune monitoring. Today, many ex vivo assays are available to be used as monitoring tools, and the selection of a robust assay that will reliably measure the frequency and/or activity of epitope-specific T cells in the peripheral circulation or patients' body fluids is critical for success. In general, recent emphasis has been on measuring tumor antigen-specific humoral and cellular responses in single cell assays. These and other DC functional assays represent a special challenge, because specimen cryopreservation and batching used in serial monitoring often interfere with cellular functions. Standardization and the selection of robust, reliable monitoring methods used in the setting of an established quality control program are the key to successful evaluations of DC-based cancer vaccines.

Theresa L. Whiteside, PhD, is a professor at the University of Pittsburgh Cancer Institute and director of the Immunologic Monitoring and Cellular Products Laboratory (IMCPL), Pittsburgh, PA, 412.624.0096, whitesidetl@upmc.edu

REFERENCES

1. Ridgway D. The first 1000 dendritic cell vaccines. Cancer Invest. 2003;21(6):873–86.

2. Cranmer LD, Trevor KT, Hersh EM. Clinical applications of dendritic cell vaccination in the treatment of cancer. Cancer Immunol Immunother. 2004;53(4):275–306.

3. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ, et al. Immunobiology of dendritic cells. Annu Rev Immunol. 2000;18:767–811.

4. Steinman RM. The dendritic cell system and its role in immunogenicity. Annu Rev Immunol. 1991;9:271–96.

5. Sabado RL, Babcock E, Kavanagh DG, Tjomsland V, Walker BD, et al. Pathways utilized by dendritic cells for binding, uptake, processing and presentation of antigens delivered from HIV-1. Eur J Immunol. 2007;Epub ahead of print, May 29.

6. Ferris RL, Whiteside TL, Ferrone S. Immune escape associated with functional defects in antigen processing machinery in head and neck cancer. Clin Cancer Res. 2007;12(13):3890–3895.

7. Sallusto F, Lanzavecchia A. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and down-regulated by tumor necrosis factor-alpha. J Exp Med. 1994;179(4):1109–18.

8. Wesa A, Kalinski P, Kirkwood JM, Tatsumi T, Storkus, WJ. Polarized type-1 dendritic cells (DC1) producing high levels of IL-12 family members rescue patient TH1-type antimelanoma CD4+ T cell responses in vitro. J Immunother. 1997;30(1):75–82.

9. Dudziak D, Kamphorst AO, Heidkamp, GF, et al. Differential antigen processing by dendritic cell subsets in vivo. Science. 2007;315(5808):107–111.

10. Liu YJ. Dendritic cell subsets and lineages, and their functions in innate and adaptive immunity. Cell. 2001;106(3):259–62.

11. Moser M, Murphy KM. Dendritic cell regulation of TH1-TH2 development. Nat Immunol. 2000;1(3):199–205.

12. Banchereau J, Schuler-Thurner B, Palucka AK, Schuler G. Dendritic cells as vectors for therapy. Cell. 2001;106(3):271–4.

13. Palucka AK, Ueno H, Connolly J, Kerneis-Norvell F, et al. Dendritic cells loaded with killed allogeneic melanoma cells can induce objective clinical responses and MART-1 specific CD8+ T-cell immunity. J Immunother. (1997) 2006;29(5):545–57.

14. Gilboa E. DC-based cancer vaccines. J Clin Invest. 2007;117(5):1195–1203.

15. Connolly NC, Whiteside TL, Wilson C, et al. Therapeutic immunization with HIV-1 peptide-loaded dendritic cells is safe and immunogenic in HIV-1-infected individuals. Clin Vaccine Immunol. 2007; In press.

16. Lu W, Arraes LC, Ferreira WT, Andrieu JM. Therapeutic dendritic-cell vaccine for chronic HIV-1 infection. Nat Med. 2004;10:1359–1365.

17. Lo J, Clare-Salzler MJ. Dendritic cell subsets and type I diabetes: focus upon DC-based therapy. Autoimmune Rev. 2006;5(6):419–23.

18. Young JW, Merad M, Hart DN. Dendritic cells in transplantation and immune-based therapies. Biol Blood Marrow Transplant. 2007;13(1):23–32.

19. Morse MA, Lyerly HK, Li Y. The role of IL-13 in the generation of dendritic cells in vitro. J Immunother. 1999;22(6):506–13.

20. Curti A, Fogli M, Ratta M, Biasco G, Tura S, Lemoli RM. Dendritic cell differentiation from hematopoietic CD34+ progenitor cells. J Biol Regul Homeost Agents. 2001;15(1):49–52.

21. Nicolette C, Healey D, Tcherepanova I, et al. Dendritic cells for active immunotherapy: Optimizing design and manufacture in order to develop commercially and clinically viable products. Vaccine. 2007; In press.

22. Zhong H, Shurin MR, Han B. Optimizing dendritic cell-based immunotherapy for cancer. Expert Rev Vaccines. 2007;6(3):333–45.

23. Sorg RV, Ozcan Z, Brefort T, Fischer J et al. Clinical-scale generation of dendritic cells in a closed system. J Immunother (1997). 2003;26(4):374–83.

24. Karp CL, Wysocka M, Ma X, Marovich M, Factor RE, Nutman T, et al. Potent suppression of IL-12 production from monocytes and dendritic cells during endotoxin tolerance. Eur J Immunol. 1998;28(10):3128–36.

25. Noorman F, Braat EA, Barrett-Bergshoeff M, Barbe E, van Leeuwen A, Lindeman J, et al. Monoclonal antibodies against the human mannose receptor as a specific marker in flow cytometry and immunohistochemistry for macrophages. J Leukoc Biol. 1997;61(1):63–72.

26. Butterfield L, Whiteside T. Development of a potency assay for human dendritic cells: IL-12p70 production. J Immunother. 2007; In press.

27. Tatsumi T, Huang J, Gooding WE, Gambotto A, et al. Intratumoral delivery of dendritic cells engineered to secrete both interleukin (IL)-12 and IL-18 effectively treats local and distant disease in association with broadly reactive TC1-type immunity. Cancer Res. 2003;63(19):6378–86.

28. Wiltrout RH, Gregorio TA, Fenton RG, Longo DL et al. Cellular and molecular studies in the treatment of murine renal cancer. Semin Oncol. 1995;22(1):9–16.

29. Elias M, van Zanten J, Hospers G, Setroikromo A, de Jong M, de Leij L, et al. Closed system generation of dendritic cells from a single blood volume for clinical application in immunotheraphy. J Clin Apher. 2005;20:197–207.

30. Tuyaerts S, Noppe SM, Corthals J, Breckpot K, Heirman C, De Greef C, et al. Generation of large numbers of dendritic cells in a closed system using cell factories. J Immunol Methods. 2002;264(1-2):135–51.

31. Proudfoot O, Apostolopoulos V, Pietersz GA. Receptor-mediated delivery of antigens to dendritic cells: anticancer applications. Mol Pharm. 2007;4(1):58-72.

32. Whiteside TL, Stanson J, Shurin MR, Ferrone S. Antigen-processing machinery in human dendritic cells: up-regulation by maturation and down-regulation by tumor cells. J Immunol. 2004;173(3):1526–34.

33. Williams MA, Porter M, Horton M, Guo J, Roman J, Williams D, et al. Ambient particulate matter directs nonclassic dendritic cell activation and a mixed TH1/TH2-like cytokine response by naive CD4+ T cells. J Allergy Clin Immunol. 2007;119(2):488–97.

34. Grohmann U, Fallarino F, Bianchi R, Belladonna ML, et al. Functional plasticity of dendritic cell subsets as mediated by CD40 versus B7 activation. J Immunol. 2003;171:2581–87.

35. Shankar G, Fourrier MS, Grevenkamp MA, Lodge PA. Validation of the COSTIM bioassay for dendritic cell potency. J Pharm Biomed Anal. 2004;36(2):285–94.

36. Whiteside T. Immune monitoring of clinical trials with biotherapies. Adv Clin Chem. 2007; In press.

37. Newton PJ, Weller IV, Katz DR, Chain BM. Autologous apoptotic T cells interact with dendritic cells, but do not affect their surface phenotype or their ability to induce recall immune responses. Clin Exp Immunol. 2003;133(1):50–8.

38. Whiteside TL. Immune suppression in cancer: Effects on immune cells, mechanisms and future therapeutic intervention. Semin Cancer Biol. 2006;16:3–15.

39. Whiteside TL. Monitoring of antigen-specific cytolytic T lymphocytes in cancer patients receiving immunotherapy. Clin Diagn Lab Immunol. 2000;7(3):327–32.

40. Gjertsen MK, Buanes T, Rosseland AR, et al. Intradermal ras peptide vaccination with granulocyte-macrophage colony-stimulating factor as adjuvant: Clinical and immunological responses in patients with pancreatic adenocarcinoma. Int J Cancer. 2001;92(3):441–50.

41. Whiteside TL, Gooding W. Immune monitoring of human gene therapy trials: potential application to leukemia and lymphoma. Blood Cells Mol Dis. 2003;31(1):63–71.

42. Prussin C, Metcalfe DD. Detection of intracytoplasmic cytokine using flow cytometry and directly conjugated anti-cytokine antibodies. J Immunol Methods. 1995;188(1):117–28.

43. Whiteside TL. ELISPOT assays. In: Nagorsen D, Marincola, FM, ed. Analyzing T Cell Responses. New York: Springer Publishers; 2005:143–57.

44. Hoffmann TK, Donnenberg VS, Friebe-Hoffmann U, et al. Competition of peptide-MHC class I tetrameric complexes with anti-CD3 provides evidence for specificity of peptide binding to the TCR complex. Cytometry. 2000;41(4):321–28.

45. Whiteside TL, Zhao Y, Tsukishiro T, Elder EM, Gooding W, Baar J. Enzyme-linked immunospot, cytokine flow cytometry, and tetramers in the detection of T-cell responses to a dendritic cell-based multipeptide vaccine in patients with melanoma. Clin Cancer Res. 2003;9(2):641–49.

46. Markovic SN, Nevala WK, Uhl CB, Celis E, McKean DJ. A reproducible method for the enumeration of functional (cytokine producing) versus non-functional peptide-specific cytotoxic T lymphocytes in human peripheral blood. Clin Exp Immunol. 2006;145(3):438–47.

47. Asai T, Storkus WJ, Whiteside TL. Evaluation of the modified ELISPOT assay for gamma interferon production in cancer patients receiving antitumor vaccines. Clin Diagn Lab Immunol. 2000;7(2):145–54.

48. Altman JD, Moss PA, Goulder PJ, et al. Phenotypic analysis of antigen-specific T lymphocytes. Science. 1996;274(5284):94–6.

49. Albers AE, Ferris RL, Kim GG, Chikamatsu K, DeLeo AB, Whiteside TL. Immune responses to p53 in patients with cancer: enrichment in tetramer+ p53 peptide-specific T cells and regulatory T cells at tumor sites. Cancer Immunol Immunother. 2005;54(11):1072–81.

50. Speiser DE, Pittet MJ, Guillaume P, et al. Ex vivo analysis of human antigen-specific CD8+ T-cell responses: quality assessment of fluorescent HLA-A2 multimer and interferon-gamma ELISPOT assays for patient immune monitoring. J Immunother. 2004;27(4):298–308.

51. Drake DR, 3rd, Ream RM, Lawrence CW, Braciale TJ. Transient loss of MHC class I tetramer binding after CD8+ T cell activation reflects altered T cell effector function. J Immunol. 2005;175(3):1507–15.

52. Miltenyi S, Muller W, Weichel W, Radbruch A. High gradient magnetic cell separation with MACS. Cytometry. 1990;11(2):231–38.

53. Okamoto Y, Nishida M. Dual-color ELISPOT assay for analyzing cytokine balance. Methods Mol Biol. 2005;302:263–72.

54. Osada T, Clay TM, Woo CY, Morse MA, Lyerly K. Dendritic cell-based immunotherapy. Int Rev Immunol. 2007;25(5):377–413.

55. De Vries IJ, Berensen MR, Lestrehuis WJ, et al. Immunomonitoring for tumor-specific T cells in delayed-type hypersensitivity skin biopsies after dendritic cell vaccination correlates with clinical outcome. J Clin Oncol. 2005;23(24):5779–87.

56. Wierecky J, Muller MR, Wirths S, et al. Immunologic and clinical responses after vaccination with peptide-pulsed dendritic cells in metastatic renal cancer patients. Cancer Res. 2006;66(11):5910–18.