PET respiratory motion correction: quo vadis?

Positron emission tomography (PET) respiratory motion correction has been a subject of great interest for the last twenty years, prompted mainly by the development of multimodality imaging devices such as PET/computed tomography (CT) and PET/magnetic resonance imaging (MRI). PET respiratory motion correction involves a number of steps including acquisition synchronization, motion estimation and finally motion correction. The synchronization steps include the use of different external device systems or data driven approaches which have been gaining ground over the last few years. Patient specific or generic motion models using the respiratory synchronized datasets can be subsequently derived and used for correction either in the image space or within the image reconstruction process. Similar overall approaches can be considered and have been proposed for both PET/CT and PET/MRI devices. Certain variations in the case of PET/MRI include the use of MRI specific sequences for the registration of respiratory motion information. The proposed review includes a comprehensive coverage of all these areas of development in field of PET respiratory motion for different multimodality imaging devices and approaches in terms of synchronization, estimation and subsequent motion correction. Finally, a section on perspectives including the potential clinical usage of these approaches is included.

Core Imaging Library - Part I: a versatile Python framework for tomographic imaging.

We present the Core Imaging Library (CIL), an open-source Python framework for tomographic imaging with particular emphasis on reconstruction of challenging datasets. Conventional filtered back-projection reconstruction tends to be insufficient for highly noisy, incomplete, non-standard or multi-channel data arising for example in dynamic, spectral and in situ tomography. CIL provides an extensive modular optimization framework for prototyping reconstruction methods including sparsity and total variation regularization, as well as tools for loading, preprocessing and visualizing tomographic data. The capabilities of CIL are demonstrated on a synchrotron example dataset and three challenging cases spanning golden-ratio neutron tomography, cone-beam X-ray laminography and positron emission tomography. This article is part of the theme issue 'Synergistic tomographic image reconstruction: part 2'.

Theranostic SPECT reconstruction for improved resolution: application to radionuclide therapy dosimetry.

BACKGROUND: SPECT-derived dose estimates in tissues of diameter less than 3× system resolution are subject to significant losses due to the limited spatial resolution of the gamma camera. Incorporating resolution modelling (RM) into the SPECT reconstruction has been proposed as a possible solution; however, the images produced are prone to noise amplification and Gibbs artefacts. We propose a novel approach to SPECT reconstruction in a theranostic setting, which we term SPECTRE (single photon emission computed theranostic reconstruction); using a diagnostic PET image, with its superior resolution, to guide the SPECT reconstruction of the therapeutic equivalent. This report demonstrates a proof in principle of this approach. METHODS: We have employed the hybrid kernelised expectation maximisation (HKEM) algorithm implemented in STIR, with the aim of producing SPECT images with PET-equivalent resolution. We demonstrate its application in both a dual 68Ga/177Lu IEC phantom study and a clinical example using 64Cu/67Cu. RESULTS: SPECTRE is shown to produce images comparable in accuracy and recovery to PET with minimal introduction of artefacts and amplification of noise. CONCLUSION: The SPECTRE approach to image reconstruction shows improved quantitative accuracy with a reduction in noise amplification. SPECTRE shows great promise as a method of improving SPECT radioactivity concentrations, directly leading to more accurate dosimetry estimates in small structures and target lesions. Further investigation and optimisation of the algorithm parameters is needed before this reconstruction method can be utilised in a clinical setting.

Estimation of optimal number of gates in dual gated 18F-FDG cardiac PET.

Gating of positron emission tomography images has been shown to reduce the motion effects, especially when imaging small targets, such as coronary plaques. However, the selection of optimal number of gates for gating remains a challenge. Selecting too high number of gates results in a loss of signal-to-noise ratio, while too low number of gates does remove only part of the motion. Here, we introduce a respiratory-cardiac motion model to determine the optimal number of respiratory and cardiac gates. We evaluate the model using a realistic heart phantom and data from 12 cardiac patients (47-77 years, 64.5 on average). To demonstrate the benefits of our model, we compared it with an existing respiratory model. Based on our study, the optimal number of gates was determined to be five respiratory and four cardiac gates in the phantom and patient studies. In the phantom study, the diameter of the most active hot spot was reduced by 24% in the dual gated images compared to non-gated images. In the patient study, the thickness of myocardium wall was reduced on average by 21%. In conclusion, the motion model can be used for estimating the optimal number of respiratory and cardiac gates for dual gating.

Rapid processing of PET list-mode data for efficient uncertainty estimation and data analysis.

In this technical note we propose a rapid and scalable software solution for the processing of PET list-mode data, which allows the efficient integration of list mode data processing into the workflow of image reconstruction and analysis. All processing is performed on the graphics processing unit (GPU), making use of streamed and concurrent kernel execution together with data transfers between disk and CPU memory as well as CPU and GPU memory. This approach leads to fast generation of multiple bootstrap realisations, and when combined with fast image reconstruction and analysis, it enables assessment of uncertainties of any image statistic and of any component of the image generation process (e.g. random correction, image processing) within reasonable time frames (e.g. within five minutes per realisation). This is of particular value when handling complex chains of image generation and processing. The software outputs the following: (1) estimate of expected random event data for noise reduction; (2) dynamic prompt and random sinograms of span-1 and span-11 and (3) variance estimates based on multiple bootstrap realisations of (1) and (2) assuming reasonable count levels for acceptable accuracy. In addition, the software produces statistics and visualisations for immediate quality control and crude motion detection, such as: (1) count rate curves; (2) centre of mass plots of the radiodistribution for motion detection; (3) video of dynamic projection views for fast visual list-mode skimming and inspection; (4) full normalisation factor sinograms. To demonstrate the software, we present an example of the above processing for fast uncertainty estimation of regional SUVR (standard uptake value ratio) calculation for a single PET scan of (18)F-florbetapir using the Siemens Biograph mMR scanner.

Detecting and estimating head motion in brain PET acquisitions using raw time-of-flight PET data.

Head motion during brain PET imaging is not uncommon and can negatively affect image quality. Motion correction techniques typically either use hardware to prospectively measure head motion, or they divide the acquisition into short fixed-frames and then align and combine these to produce a motion free image. The aim of this work was to retrospectively detect when motion occurred in PET data without the use of motion detection hardware, and then align the frames defined by these motion occurrences. We describe two methods that use either principal component analysis or the motion induced spatial displacements over time to detect motion in raw time-of-flight PET data. The points in time of motion then define the temporal boundaries of frames which are reconstructed without attenuation correction, aligned and combined. Phantom and [18F]-Fallypride patient acquisitions were used to validate and evaluate these approaches, which were compared with motion estimation using 60 s fixed-frames. Both methods identified all motion occurrences in phantom data, and unlike the fixed-frame approach did not exhibit intra-frame motion. With patient acquisitions, images corrected with the motion detection methods increased the average image sharpness by the same amount as the fixed-frame approach, but reduced the number of reconstructions and registrations by a factor of 3.4 on average. Detecting head motion in raw PET data alone is possible, allowing retrospective motion estimation of any listmode brain PET acquisition without additional hardware, subsequently decreasing data processing and potentially reducing intra-frame motion.

Extracting a respiratory signal from raw dynamic PET data that contain tracer kinetics.

Data driven gating (DDG) methods provide an alternative to hardware based respiratory gating for PET imaging. Several existing DDG approaches obtain a respiratory signal by observing the change in PET-counts within specific regions of acquired PET data. Currently, these methods do not allow for tracer kinetics which can interfere with the respiratory signal and introduce error. In this work, we produced a DDG method for dynamic PET studies that exhibit tracer kinetics. Our method is based on an existing approach that uses frequency-domain analysis to locate regions within raw PET data that are subject to respiratory motion. In the new approach, an optimised non-stationary short-time Fourier transform was used to create a time-varying 4D map of motion affected regions. Additional processing was required to ensure that the relationship between the sign of the respiratory signal and the physical direction of movement remained consistent for each temporal segment of the 4D map. The change in PET-counts within the 4D map during the PET acquisition was then used to generate a respiratory curve. Using 26 min dynamic cardiac NH3 PET acquisitions which included a hardware derived respiratory measurement, we show that tracer kinetics can severely degrade the respiratory signal generated by the original DDG method. In some cases, the transition of tracer from the liver to the lungs caused the respiratory signal to invert. The new approach successfully compensated for tracer kinetics and improved the correlation between the data-driven and hardware based signals. On average, good correlation was maintained throughout the PET acquisitions.

Interference with PD-L1/PD-1 co-stimulation during antigen presentation enhances the multifunctionality of antigen-specific T cells.

The release of cytokines by T cells strongly defines their functional activity in vivo. The ability to produce multiple cytokines has been associated with beneficial immune responses in cancer and infectious diseases, while their progressive loss is associated with T-cell exhaustion, senescence and anergy. Consequently, strategies that enhance the multifunctional status of T cells are a key for immunotherapy. Dendritic cells (DCs) are professional antigen presenting cells that regulate T-cell functions by providing positive and negative co-stimulatory signals. A key negative regulator of T-cell activity is provided by binding of programmed death-1 (PD-1) receptor on activated T cells, to its ligand PD-L1, expressed on DCs. We investigated the impact of interfering with PD-L1/PD-1 co-stimulation on the multifunctionality of T cells, by expression of the soluble extracellular part of PD-1 (sPD-1) or PD-L1 (sPD-L1) in human monocyte-derived DCs during antigen presentation. Expression, secretion and binding of these soluble molecules after mRNA electroporation were demonstrated. Modification of DCs with sPD-1 or sPD-L1 mRNA resulted in increased levels of the co-stimulatory molecule CD80 and a distinct cytokine profile, characterized by the secretion of IL-10 and TNF-α, respectively. Co-expression in DCs of sPD-1 and sPD-L1 with influenza virus nuclear protein 1 (Flu NP1) stimulated Flu NP1 memory T cells, with a significantly higher number of multifunctional T cells and increased cytokine secretion, while it did not induce regulatory T cells. These data provide a rationale for the inclusion of interfering sPD-1 or sPD-L1 in DC-based immunotherapeutic strategies.

In vitro generation of mature, naive antigen-specific CD8(+) T cells with a single T-cell receptor by agonist selection.

Peripheral blood T cells transduced with a tumor-specific T-cell receptor (TCR) face problems of auto-reactivity and lack of efficacy caused by cross-pairing of exogenous and endogenous TCR chains, as well as short term in vivo survival due to activation and growth factor-induced differentiation. We here studied an alternative strategy for the efficient generation of naive CD8(+) T cells with a single TCR. TCR-transduced human postnatal thymus-derived and adult mobilized blood-derived hematopoietic progenitor cells (HPCs) were differentiated to CD4(+)CD8(+) double-positive T cells using OP9-Delta-like 1 (OP9-DL1) cultures. Addition of the agonist peptide induced double positive cells to cross-present the peptide, leading, in the absence of co-stimulation, to cell cycle arrest and differentiation into mature CD8(+) T cells. Comprehensive phenotypic, molecular and functional analysis revealed the generation of naive and resting CD8(+) T cells through a process similar to thymic positive selection. These mature T cells show a near complete inhibition of endogenous TCRA and TCRB rearrangements and express high levels of the introduced multimer-reactive TCR. Upon activation, specific cytokine production and efficient killing of tumor cells were induced. Using this strategy, large numbers of high-avidity tumor-specific naive T cells can be generated from readily available HPCs without TCR chain cross-pairing.

A phase IB study on intravenous synthetic mRNA electroporated dendritic cell immunotherapy in pretreated advanced melanoma patients.

BACKGROUND: Autologous monocyte-derived dendritic cells (DCs) electroporated with synthetic messenger RNA (mRNA) encoding a CD40 ligand, a constitutively active Toll-like receptor 4 and CD70, together with mRNA encoding fusion proteins of a human leukocyte antigen (HLA)-class II targeting signal (DC-LAMP) and a melanoma-associated antigen (MAA); either MAGE-A3, MAGE-C2, tyrosinase or gp100) (TriMixDC-MEL) are superiorly immunogenic. PATIENTS AND METHODS: In this phase IB clinical trial, 24 million viable DCs were administered by four biweekly combined intradermal (id) and intravenous (iv) administrations, and a fifth administration on week 16. The number of iv-administered DCs was escalated in four sequentially treated cohorts. Immune responses were assessed by analysis of antigen specificity of blood-derived T-cells and skin infiltrating lymphocytes (SKILs). RESULTS: Fifteen patients with pretreated advanced melanoma tolerated administration of TriMixDC-MEL well. Two patients achieved a complete response and two patients a partial response. All objective responders are progression-free after a follow-up of, respectively, 24+, 28+, 33+, and 34+ months. Post-therapy antigen-specific SKILs were documented in 6 of 12 patients, and antigen-specific CD8(+) T-cells were detected in the blood of 4 of 5 patients. CONCLUSIONS: Cellular immunotherapy with TriMixDC-MEL is safe and immunogenic. Antitumor activity with durable disease control is observed across the investigated iv-dose levels. CLINICALTRIALSGOV IDENTIFIER: NCT01066390.

Downregulation of Stat3 in melanoma: reprogramming the immune microenvironment as an anticancer therapeutic strategy.

Persistent activation of the transcription factor, signal transducer and activator of transcription 3 (Stat3) has been shown to mediate several oncogenic features in many types of cancers, including melanoma. In this study, we investigated whether lentiviral (LV) delivery of Stat3-targeting short hairpin RNA (shRNA; LV-shStat3) to K1735-C4 melanoma cells modulates antitumor immunity. Three shStat3 sequences, starting at the position 446, 830 and 1412, were cloned into a mir30 cassette. A shRNA with scrambled sequence served as a control. Transduction with LV-shStat3 resulted in downregulation of Stat3 in vitro. The latter coincided with low cell viability, a reduced expression of survivin and matrix metalloproteinase (MMP)-2. A single injection of LV-shStat3 in K1735-C4 tumors efficiently downregulated Stat3 in vivo and resulted in reduction of both vascular endothelial growth factor secretion and in myeloid-derived suppressor cell (MDSC) numbers. In contrast, we observed an increase in interleukin-6 and interferon-γ secretion, mature dendritic cells (DCs) and CD8(+) T cells. Both DCs and CD8(+) T cells displayed enhanced activity, whereas granulocytic MDSCs lost their suppressive capacity upon Stat3 downregulation. Importantly, a single injection of LV-shStat3 was sufficient to reduce tumor growth, hence prolong survival of tumor-bearing mice. These data demonstrate that Stat3 downregulation in melanoma reinvigorates existing antitumor immunity.

The effect of regularization in motion compensated PET image reconstruction: a realistic numerical 4D simulation study.

Following continuous improvement in PET spatial resolution, respiratory motion correction has become an important task. Two of the most common approaches that utilize all detected PET events to motion-correct PET data are the reconstruct-transform-average method (RTA) and motion-compensated image reconstruction (MCIR). In RTA, separate images are reconstructed for each respiratory frame, subsequently transformed to one reference frame and finally averaged to produce a motion-corrected image. In MCIR, the projection data from all frames are reconstructed by including motion information in the system matrix so that a motion-corrected image is reconstructed directly. Previous theoretical analyses have explained why MCIR is expected to outperform RTA. It has been suggested that MCIR creates less noise than RTA because the images for each separate respiratory frame will be severely affected by noise. However, recent investigations have shown that in the unregularized case RTA images can have fewer noise artefacts, while MCIR images are more quantitatively accurate but have the common salt-and-pepper noise. In this paper, we perform a realistic numerical 4D simulation study to compare the advantages gained by including regularization within reconstruction for RTA and MCIR, in particular using the median-root-prior incorporated in the ordered subsets maximum a posteriori one-step-late algorithm. In this investigation we have demonstrated that MCIR with proper regularization parameters reconstructs lesions with less bias and root mean square error and similar CNR and standard deviation to regularized RTA. This finding is reproducible for a variety of noise levels (25, 50, 100 million counts), lesion sizes (8 mm, 14 mm diameter) and iterations. Nevertheless, regularized RTA can also be a practical solution for motion compensation as a proper level of regularization reduces both bias and mean square error.

Lentiviral vectors: a versatile tool to fight cancer.

Over the years, there has been an exponential increase in the number of gene therapy approaches that are under investigation for the treatment of cancer. This can be attributed to our growing understanding of the molecular mechanisms that contribute to the onset and maintenance of cancer as well as to the development of gene delivery vectors. In this review, we will focus on the use of lentiviral vectors (LVs) in immuno gene therapy of cancer, as these efficacious gene delivery vehicles have come to the fore front because of their many attractive features. LVs have been successfully applied to generate potent dendritic cell based anti-cancer vaccines and to deliver cancer-specific receptors to T-cells. Moreover, LVs are under investigation for the modulation of cancer cells. We will describe various strategies of this 'genuine' cancer gene therapy, amongst which transfer of suicide genes, modulation of pro- and anti-apoptotic molecules, strategies to optimize chemo- and radiotherapy, expression of molecules that affect angiogenesis or affect the immunogenicity of tumor cells. These will be discussed in view of our current knowledge of tumor immunology. Finally we will discuss some important issues and future directions to push the field forward.

Sequence evolution and escape from specific immune pressure of an HIV-1 Rev epitope with extensive sequence similarity to human nucleolar protein 6.

Antigen-specific immunity is crucially important for containing viral replication in human immunodeficiency virus (HIV)-1-infected hosts. Several epitopes have been predicted for the early expressed HIV-1 proteins Tat and Rev, but few have been studied in detail. We characterized the human leukocyte antigen (HLA)-B44-restricted Rev epitope EELLKTVRL (EL9) in an HIV-1-infected subject treated with antiretroviral therapy. Interestingly, a high sequence similarity was found between the EL9 epitope and the human nucleolar protein 6 (NOL6). However, this similarity does not seem to impede immunogenicity as CD8(+) T-cells, previously stimulated with EL9-pulsed dendritic cells, were able to specifically recognize the HIV-1 Rev epitope without cross-recognizing the human self-antigen NOL6. After the subject interrupted antiretroviral therapy and virus rebounded, mutations within the EL9 epitope were identified. Although the emerging mutations resulted in decreased or abolished T-cell recognition, they did not impair Rev protein function. Mutations leading to escape from T-cell recognition persisted for up to 124 weeks following treatment interruption. This study shows that the HLA-B44-restricted Rev CD8(+) T-cell epitope EL9 is immunogenic notwithstanding its close resemblance to a human peptide. The epitope mutates as a consequence of dynamic interaction between T-cells and HIV-1. Clinical status, CD4(+) T-cell count and viral load remained stable despite escape from T-cell recognition.

Development of the Nanobody display technology to target lentiviral vectors to antigen-presenting cells.

Lentiviral vectors (LVs) provide unique opportunities for the development of immunotherapeutic strategies, as they transduce a variety of cells in situ, including antigen-presenting cells (APCs). Engineering LVs to specifically transduce APCs is required to promote their translation towards the clinic. We report on the Nanobody (Nb) display technology to target LVs to dendritic cells (DCs) and macrophages. This innovative approach exploits the budding mechanism of LVs to incorporate an APC-specific Nb and a binding-defective, fusion-competent form of VSV.G in the viral envelope. In addition to production of high titer LVs, we demonstrated selective, Nb-dependent transduction of mouse DCs and macrophages both in vitro and in situ. Moreover, this strategy was translated to a human model in which selective transduction of in vitro generated or lymph node (LN)-derived DCs and macrophages, was demonstrated. In conclusion, the Nb display technology is an attractive approach to generate LVs targeted to specific cell types.

Fighting with the enemy's weapons? the role of costimulatory molecules in HIV.

HIV infection is characterized by a number of abnormalities in several components of the immune system. For example, during HIV infection, a massive decrease of CD4(+) T cells is observed, as well as a progressive depletion of naïve CD8(+) T cells. Furthermore, elevated numbers of apoptotic B and T cells are present in HIV-infected patients, and a systemic immune activation results in T-cell exhaustion. Finally, HIV infection is characterized by the presence of functionally impaired dendritic cells, with decreased expression of maturation markers, decreased secretion of cytokines and defects in antigen processing and presentation. All these characteristics result in the occurrence of non-functional cytotoxic T lymphocytes, that fail to control HIV-replication in most individuals during progressive disease. Costimulatory and co-inhibitory molecules are involved in the activation, differentiation and survival of several cell-types of the immune system. Each costimulatory receptor (generally expressed on effector cells) can conjugate with one or more specific ligands (expressed on antigen-presenting cells), which leads to an activation of intracellular signaling pathways inside the cells on which they are expressed. HIV infection is characterized by an aberrant expression of these molecules on cells of the immune system. Many of the immune deficiencies mentioned in the previous paragraph can be explained by abnormal expression of costimulatory molecules, and could consequently be overcome by interfering with their interactions. In this review, we give an overview of the functions and expression patterns of the receptor/ligand pairs of the tumor necrosis factor and the B7 super-families of costimulatory and co-inhibitory molecules in HIV-infected patients. We will also discuss possibilities for manipulating their signaling as a therapeutic anti-HIV tool.

Image-based point spread function implementation in a fully 3D OSEM reconstruction algorithm for PET.

The interest in positron emission tomography (PET) and particularly in hybrid integrated PET/CT systems has significantly increased in the last few years due to the improved quality of the obtained images. Nevertheless, one of the most important limits of the PET imaging technique is still its poor spatial resolution due to several physical factors originating both at the emission (e.g. positron range, photon non-collinearity) and at detection levels (e.g. scatter inside the scintillating crystals, finite dimensions of the crystals and depth of interaction). To improve the spatial resolution of the images, a possible way consists of measuring the point spread function (PSF) of the system and then accounting for it inside the reconstruction algorithm. In this work, the system response of the GE Discovery STE operating in 3D mode has been characterized by acquiring (22)Na point sources in different positions of the scanner field of view. An image-based model of the PSF was then obtained by fitting asymmetric two-dimensional Gaussians on the (22)Na images reconstructed with small pixel sizes. The PSF was then incorporated, at the image level, in a three-dimensional ordered subset maximum likelihood expectation maximization (OS-MLEM) reconstruction algorithm. A qualitative and quantitative validation of the algorithm accounting for the PSF has been performed on phantom and clinical data, showing improved spatial resolution, higher contrast and lower noise compared with the corresponding images obtained using the standard OS-MLEM algorithm.

Lentiviral vectors for cancer immunotherapy: transforming infectious particles into therapeutics.

Lentiviral vectors have emerged as promising tools for both gene therapy and immunotherapy purposes. They exhibit several advantages over other viral systems in that they are less immunogenic and are capable of transducing a wide range of different cell types, including dendritic cells (DC). DC transduced ex vivo with a whole range of different (tumor) antigens were capable of inducing strong antigen-specific T-cell responses, both in vitro and in vivo. Recently, the administration of lentiviral vectors in vivo has gained substantial interest as an alternative method for antigen-specific immunization. This method offers a number of advantages over DC vaccines as the same lentivirus can in principle be used for all patients resulting in a significantly reduced cost and requirement for considerably less expertise for the generation and administration of lentiviral vaccines. By selectively targeting lentiviral vectors to, or restricting transgene expression in certain cell types, selectivity, safety and efficacy can be further improved. This review will focus on the use of direct administration of lentiviral vectors encoding tumor-associated antigens (TAA) for the induction of tumor-specific immune responses in vivo, with a special focus on problems related to the generation of large amounts of highly purified virus and specific targeting of antigen-presenting cells (APC).

Induction of antigen-specific CD8+ cytotoxic T cells by dendritic cells co-electroporated with a dsRNA analogue and tumor antigen mRNA.

The maturation state of dendritic cells (DCs) is an important determinant for the initiation and regulation of adaptive immune responses. In this study, we wanted to assess whether functional activation of human monocyte-derived DCs can be achieved by electroporation of an activation signal in the form of double-stranded (ds) RNA and whether simultaneous electroporation of the dsRNA with tumor antigen encoding mRNA can lead to the induction of a cytotoxic T-lymphocyte (CTL) response. Electroporation of immature DCs with poly(I:C(12)U), a dsRNA analogue, resulted in phenotypic as well as functional changes, indicative of DC maturation. Co-electroporation of DCs with both poly(I:C(12)U) and Melan-A/MART-1 encoding mRNA induced strong anti-Melan-A/MART-1 CD8(+) T-cell responses in vitro. Higher numbers of Melan-A/MART-1-specific CTLs were consistently obtained with poly(I:C(12)U)-activated DCs compared to DCs matured in the presence of an inflammatory cytokine cocktail. These results indicate that DC co-electroporation with both dsRNA and tumor antigen encoding mRNA induces fully activated and antigen-loaded DCs that promote antigen-specific CTL responses and may provide the basis for future immunotherapeutic strategies.

Induction of effective therapeutic antitumor immunity by direct in vivo administration of lentiviral vectors.

Ex vivo lentivirally transduced dendritic cells (DC) have been described to induce CD8+ and CD4+ T-cell responses against various tumor-associated antigens (TAAs) in vitro and in vivo. We report here that direct administration of ovalbumin (OVA) encoding lentiviral vectors caused in vivo transduction of cells that were found in draining lymph nodes (LNs) and induced potent anti-OVA cytotoxic T cells similar to those elicited by ex vivo transduced DC. The cytotoxic T-lymphocyte (CTL) response following direct injection of lentiviral vectors was highly effective in eliminating target cells in vivo up to 30 days after immunization and was efficiently recalled after a boost immunization. Injection of lentiviral vectors furthermore activated OVA-specific CD4+ T cells and this CD4 help was shown to be necessary for an adequate primary and memory CTL response. When tested in therapeutic tumor experiments with OVA+ melanoma cells, direct administration of lentiviral vectors slowed down tumor growth to a comparable extent with the highest dose of ex vivo transduced DC. Taken together, these data indicate that direct in vivo administration of lentiviral vectors encoding TAAs has strong potential for anticancer vaccination.