Skip to main content
Insights into Imaging
Download PDF
  • Critical Review
  • Open access
  • Published:

The use of PET/MRI in radiotherapy

Insights into Imaging volume 15, Article number: 63 (2024) Cite this article

Abstract

Positron emission tomography/magnetic resonance imaging (PET/MRI) is a hybrid imaging technique that quantitatively combines the metabolic and functional data from positron emission tomography (PET) with anatomical and physiological information from MRI. As PET/MRI technology has advanced, its applications in cancer care have expanded. Recent studies have demonstrated that PET/MRI provides unique advantages in the field of radiotherapy and has become invaluable in guiding precision radiotherapy techniques. This review discusses the rationale and clinical evidence supporting the use of PET/MRI for radiation positioning, target delineation, efficacy evaluation, and patient surveillance.

Critical relevance statement This article critically assesses the transformative role of PET/MRI in advancing precision radiotherapy, providing essential insights into improved radiation positioning, target delineation, efficacy evaluation, and patient surveillance in clinical radiology practice.

Key points

• The emergence of PET/MRI will be a key bridge for precise radiotherapy.

• PET/MRI has unique advantages in the whole process of radiotherapy.

• New tracers and nanoparticle probes will broaden the use of PET/MRI in radiation.

• PET/MRI will be utilized more frequently for radiotherapy.

Graphical Abstract

Introduction

Radiotherapy is a standard component of care for many cancer patients. With the advancement of radiotherapy technology, three-dimensional conformal radiotherapy (3D-CRT), intensity modulated radiation therapy (IMRT), stereotactic radiotherapy (SRT), and tomotherapy have arisen and dramatically enhanced the prognosis of cancer patients [1, 2]. CT is widely used in radiation planning due to its ability to provide high-resolution anatomical information, while MRI is valued for its excellent soft tissue contrast [3, 4]. Nevertheless, standard imaging techniques only reveal morphological alterations, which is insufficient for precise radiation planning. To overcome this limitation, researchers have been exploring the use of multimodal imaging to enhance its precision. One promising avenue is the integration of multi-parametric PET/MRI into one-stop-shop radiotherapy (RT) planning workflow (Fig. 1).

Fig. 1
figure 1

Comparison of PET/MRI for radiotherapy procedures with conventional radiotherapy procedures

Full size image

PET/MRI is a hybrid imaging technique that quantitatively combines metabolic functional data from PET with anatomical and physiological information from MRI [5, 6]. Due to its high soft tissue contrast, PET/MRI may be extremely accurate in T-staging of cancer, such as head and neck, abdominal, and pelvic tumors. Recent studies have begun to show that it is also valuable in treating lymphoma, liver, and breast cancer [7,8,9]. When it comes to finding lymph node metastases and faraway metastases, PET/MRI is also superior to CT, MRI, and PET/CT, as PET/MRI can provide dynamic enhanced imaging, diffusion-weighted imaging (DWI), and signal strength (SI) assessment that can display non-fluorodeoxyglucose (non-FDG) uptake lesions and avoid misdiagnosis of sites with physiologic FDG uptake, such as the adrenal glands and mucous membranes [10,11,12,13]. Thus, given the consideration of the advantages mentioned, the efficacy of PET/MRI in tumor diagnosis and staging has been the subject of extensive research [14,15,16]. Recently, increasing researches indicate that PET/MRI can also be utilized in radiotherapy and can further improve the accuracy of radiation therapy [17,18,19,20,21]. This review summarizes the available literature on the utilization of PET/MRI for radiation placement, target delineation, efficacy evaluation, and patient surveillance (Table 1).

Table 1 Comparison of PET/MRI and other conventional image-base modality in radiotherapy
Full size table

Radiotherapy positioning

Accurate radiotherapy positioning is a critical component of the pre-radiotherapy preparation process. The advent of 18F-fluorodeoxyglucose ([18F]-FDG) PET/MRI has provided clinicians with more precise and comprehensive imaging data, enhancing the effectiveness of patient positioning for radiotherapy. However, alongside this technology, specialized positioning devices have been developed and introduced which, while improving precision, also present new challenges in terms of integration and optimization.

Specific equipment and material

In addition to the standard diagnostic [18F]-FDG PET/MRI, the [18F]-FDG PET/MRI used in radiotherapy incorporates specialized equipment and a method for external-internal reference point positioning (Table 2). The radiation equipment consists of a flat table and a patient positioning device. As for the treatment bed, unlike the curved one used in diagnostic MR, it is flat in [18F]-FDG PET/MRI for radiotherapy. This is because, on the one hand, [18F]-FDG PET/MRI images need to be aligned with CT images for accurate dose calculation. In CT, the bed used for treatment is flat. If a curved treatment bed is still used, the image quality will be affected. On the other hand, patients need to be placed in the same way every time they get radiotherapy through fixed devices, which are often installed on a flat structure [22]. In reference point positioning systems, isocentric placement is mostly done with the help of laser systems [21], where markers are made on patients’ skin to ensure accurate repositioning. However, when the positioning fixture is included, the anterior receiver coil of the diagnostic machine is not compatible with the radiation equipment. Thus, a radiotherapy-specific coil and coil holder were also introduced. In this way, the near-coil effect, that the signal increases as the distance from the coil decreases, was minimized [18]. However, due to technical restrictions, the whole-body coil holder may cause MR image distortion and truncation. To restore truncations, PET-based and purely MR-based methods have been introduced [21, 23, 24].

Table 2 Comparison of diagnostic PET/MRI and PET/MRI simulator for radiotherapy positioning
Full size table

Materials in radiotherapy equipment are also critical as they directly impact the quality of imaging. Existing materials for flat tabletops are primarily composed of carbon fiber which offers minimal photon attenuation. However, the conductive nature of carbon fiber can generate surface eddy currents, leading to the production of image artifacts and signal voids, making it incompatible with MRI. Subsequently, flat tabletops made of glass fibers were introduced; however, glass fibers significantly attenuate the PET signal, leading to artifacts and increased quantization errors in PET images, rendering them incompatible with PET [25,26,27]. Based on this, materials combining a plastic sandwich with a foam core have been developed, which are designed to have reduced and homogeneous photon attenuation, making them compatible with both MRI and PET [21] (details shown in Fig. 2). For the MR coil, because the gamma-ray attenuation of materials is closely tied to the electron density distribution of those materials, the use of electron-dense materials is not an option [28]. Building on this principle, the use of thin plastic shells, thin copper wire coils, and lightweight coil technologies are being explored to reduce PET attenuation and enhance image quality [29, 30].

Fig. 2
figure 2

The difference between image quality obtained from different types of tabletop materials using (a) carbon fiber, (b) glass fiber, and (c) plastic sandwich with foam core

Full size image

Effect on image quality

With the introduction of specific equipment and material for [18F]-FDG PET/MRI, the image quality of [18F]-FDG PET/MRI may be affected. The signal-to-noise ratio (SNR) is a critical metric for evaluating the image quality of MRI. Witoszynskyj et al. reported mean SNR values at the slice through the center of the phantom were 98.0 ± 27.1 with a flat table and 98.3 ± 27.0 without it, suggesting that the flat table has a negligible effect on SNR [31]. In addition, as positioning devices increase the distance between the patient and the coil, the total SNR of MRI pictures was reduced by 25% compared with the usual setup [21]. Winter et al. observed a slight reduction in image SNR with the radiotherapy device compared to diagnostic [18F]-FDG PET/MRI. However, the target volume remained consistent with that delineated by MRI, indicating that the reduction in SNR may not pose a limitation in RT planning [17]. However, improved image quality is still required for further precision radiotherapy. Although the SNR can be improved in several ways, the process of doing so often comes with other problems. For instance, by changing the MR parameters, such as raising the signal average [32], lowering the acceleration, or shortening the time of echo (TE), the SNR can be increased, but the acquisition time will also be increased [17]. Additionally, an adjustable coil bridge was developed to solve the problem of patient-coil distance, and while moving the coil closer to the patient improves image quality, positioning repeatability may become more difficult [33]. Boosting spatial resolution, which defines the picture’s capacity to portray fine structures, is another technique for improving MRI image quality, but care must be given not to impair SNR due to the inverse relationship between spatial resolution and SNR [17]. Recently, a new noise reduction and reconstruction technique using deep learning reconstruction, which only changes the SNR, has emerged, but its accuracy needs to be further verified [34].

In PET imaging, objects within the field of view (FOV) of the PET detector can attenuate and scatter PET photons, leading to reduced image quality. Thus, thin patient positioning devices, such as thermoplastic masks, were reported and found to have a negligible effect on PET quantification [21]. Witoszynskyj et al. demonstrated, using a phantom, that image activity near the flat tabletop side was underestimated, but this can be rectified by employing attenuation correction [31]. Based on phantom and human data, Paulus et al. predicted that accounting for the coil holder, including the radiofrequency (RF) coil, in the attenuation correction would reduce the deviation from 13.8 to 0.8% [21].

Attenuation correction (AC)

[18F]-FDG PET/MRI positioning images, unlike diagnostic images that only need to provide qualitative information, will subsequently be used for dose calculations in radiation therapy and thus require an accurate AC map. The introduction of coil fixation devices in radiation equipment mitigates the problem of AC map error caused by the unfixed position and geometric shape of the flexible coil during scanning [32, 35, 36]. But another key issue arises in [18F]-FDG PET/MRI, different from PET/CT, MRI cannot provide photon attenuation data that can be directly used for AC of PET. To resolve this issue, different proposals have been suggested, including MR-based image segmentation, atlas-based AC, and substitute CT (s-CT) [37,38,39,40]. MR-based image segmentation divides the attenuated images into four categories, but bone segmentation is not performed due to the low intensity and non-specific signals of cortical bone on MRI [37]. Atlas-based AC can take into account cortical bone. However, the registration of the whole-body image to the atlas is prone to errors due to the different stiffness of the chest and abdomen and the head [41, 42]. s-CT correlates the voxel values of CT with MRI images, which can turn the MRI image into s-CT, and automatically generates attenuation maps with more fine intervals than those generated by segmentation [40]. Recently, artificial intelligence (AI) employing deep learning convolutional neural networks (CNNs) is emerging as an alternative to traditional AC methods owing to its speed, accuracy, and robustness [43]. In addition, different AC methods were generated according to the characteristics of different components. Rigid hardware components, like coil holders and flat tabletops, can undergo AC using CT-based 3D attenuation maps or 68Ge/68Ga projection scanning attenuation maps, due to their fixed positions and strict geometry [44]. In contrast, AC for flexible coils can be achieved using MRI with ultra-short echo time (MRI-UTE) sequences. However, this method has very limited visualization and provides insufficient information on the coil components for accurate AC [44, 45]. The marker-based correction method, automatically determining the position of surface coils using markers, was proposed for accurate AC [46]. Moreover, Lindemann et al. proposed a method based on computer-aided design (CAD), which generated AC-map without artifacts and improved spatial resolution [31, 32, 46, 47].

Once the attenuation map is obtained, it is vital to assess its repeatability during the radiation repositioning process. Since the [18F]-FDG PET/MRI flat tabletop remains in a fixed position, only a single registration is necessary. However, as the coil holder undergoes repeated installation and removal, it is essential to confirm the accuracy of the attenuation map during repositioning [21]. Paulus et al. evaluated the accuracy of the coil holder during multiple repositionings using a phantom and an active 68Ge rod source. They found that there were only slight deviations in accuracy when the RF coil holder was repeatedly moved without altering the phantom’s position, suggesting that the attenuation map of the coil holder is reliable for repositioning purposes [21].

MRI protocols for RT positioning

In the context of RT localization using MRI, the selection of sequences is a crucial step in ensuring both efficiency and accuracy. Typically, T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and DWI sequences are chosen. T1WI and T2WI offer essential anatomical information for registration and target delineation. DWI, providing valuable metabolic insights, aids in refining tumor boundaries and assessing lymph node involvement [48]. Recently, in order to further optimize the scanning time of MRI sequences, and to improve patient compliance while ensuring accurate clinical information, different sequences are recommended for different lesions [48,49,50,51,52,53,54]. For instance, inversion recovery gradient echo (IR-GRE) is commonly used for T1WI in brain tumors, while turbo spin-echo (TSE) and fluid-attenuated inversion recovery (FLAIR) imaging are frequently employed for T2WI [51, 52]. In areas with abundant soft tissues like the head and neck, breast, abdomen, and pelvis, fat saturation technology is often applied for improved tumor and organ visualization [48, 53, 54].

Radiotherapy planning

With the introduction of the IMRT, the significance of accurate tumor volumes is re-emphasized. Since [18F]-FDG PET/MRI can concurrently capture both tissue morphology and tumor metabolism information, it can accurately distinguish tumor tissue from neighboring tissues, making it valuable for accurate target identification in radiotherapy [55].

Target delineation

Gross tumor volume (GTV) is commonly outlined using the tumor morphology visualized on MRI. However, integrating tumor biology information from PET can mitigate the risk of marginal loss, potentially impacting GTV changes. Zhang et al. observed an increasing difference between GTV-MRI and GTV-PET with growing tumor volumes [56, 57]. Notably, while around 90% of GTV-PET overlapped with GTV-MRI, 10% of the tumor and lymph node volume was exclusively identified by PET, emphasizing the value of a combined [18F]-FDG PET/MRI approach in RT planning [58]. The emergence of PET/MRI has great hope to achieve accurate target delineation.

The effect of PET/MRI on GTV has been studied through the observation of the differences among GTV-MRI, GTV-PET, and GTV-PET/MRI [55, 58, 59]. Using GTV-MRI as a reference, Zhang et al. evaluated the differences between GTV-PET and GTV-[18F]-FDG PET/MRI in the delineation of colorectal liver metastases and found that GTV-[18F]-FDG PET/MRI had the highest tumor volume [55]. In contrast, Mahase’s work came to the opposite [59], where they found that the average volume of GTV-[68Ga]-DOTATATE PET/MRI was lower than that of GTV-MRI in patients with intracranial meningiomas. This discrepancy may be attributed to the utilization of specific tracers in PET, which enable the detection of microscopic tumor activity even in the absence of morphological changes. As a result, the obtained GTV will expand [59, 60]. Conversely, [68Ga]-DOTATOC PET/MRI allows for the visualization of the boundary between the tumor and normal tissue, as well as the necrotic parts of the tumor resulting from adjuvant treatment, which can lead to a reduced estimation of GTV [61]. Furthermore, Mahase et al. observed that the use of [68Ga]-DOTATATE PET/MRI, which can result in a reduction of GTV, consequently leading to a decrease in both clinical tumor volume (CTV)—comprising the GTV and potentially invaded tissues and planning target volume (PTV), which is derived from the expansion of CTV [59]. In addition, in the lymph node metastasis delineation, the nodal GTV (GTVnd) delineated based on PSMA PET/MRI contains metastatic pelvic lymph nodes more accurately than MRI [62]. However, Liu et al. showed that the CTVs of PSMA PET/MRI and MRI are similar in the target delineation of pelvic lymph node metastasis [62], where the distinction brought about by GTVnd is not apparent, possibly because CTV encompasses a larger range than GTVnd. Moreover, the pathological analysis also confirmed the advantage of PET/MRI in the target delineation, Zhang et al. proposed that PET/MRI is more adept at mapping GTV by comparing the Dice Similarity Coefficient (DSC) of PET/MRI and PET. Additionally, they demonstrated that the longest tumor length measured by PET/MRI correlated well with the longest tumor length measured by histopathological analysis, providing robust evidence for the superior accuracy of GTV delineation by PET/MRI [60, 62]. Figure 3 illustrates the difference between [18F]-FDG PET/MRI and other traditional imaging methods for GTV delineation. However, although the advantages of PET/MRI in target delineation have been identified, the impact of PET/MRI on treatment response and survival has not been determined, and further exploration is needed in the future.

Fig. 3
figure 3

Sixty nine-year-old female with nasopharynx cancers. a The blue line represents gross tumor volume (GTV)-PET/CT. b The pink line represents GTV-PET/MRI. c The red line represents GTV-CT. d The green line represents GTV-MRI. e The yellow line represents GTV-PET

Full size image

In addition to the abovementioned effects of PET/MRI on the target area, manual delineation of tumor volume introduces considerable potential for inter-observer variability [63], which can be mitigated through the application of standardized segmentation techniques. Co-segmentation based on [18F]-FDG PET/MRI has garnered significant interest in recent years [64,65,66]. Leibfarth et al. found that the tumor volume delineated manually by the observer was comparable to that automatically delineated by the [18F]-FDG PET/MRI method, indicating the viability of [18F]-FDG PET/MRI co-segmentation [66]. In addition, [18F]-FDG PET/MRI can provide more stable imaging [55, 67]. Cavaliere et al. demonstrated a high degree of agreement between two groups of observers in the anatomical localization and extent of lesions when using [18F]-FDG PET/MRI [68]. Moreover, the DSC of the target volume delineated by different observers on [18F]-FDG PET/MRI is greater than that on MRI, demonstrating a higher degree of overlap and less degree of discrimination. To address observer inconsistency and enhance accuracy, some studies propose the use of an adaptive threshold level (aTL) derived from individual maximum standardized uptake values (SUVmax) [58, 69,70,71]. It is important to note that the appropriate threshold is not a fixed value due to the low spatial resolution of PET and the existing partial volume effect. Published articles indicate that in lung cancer, thresholds based on a percentage of SUVmax range from 15 to 50% [72]. Alfano et al. revealed that the sensitivity and specificity could be improved using a 67% SUVmax threshold and 81% SUVmax, respectively [73]. In addition, Dietlein et al. observed that thresholds may vary depending on the chosen radiotracer [74]. Research is needed to further explore whether the threshold is related to additional tumor-related factors such as hypoxia, proliferation, and histological differences. Despite ongoing innovations [65, 66, 75], errors in automatically delineated target areas may still occur due to imprecise tumor boundaries, necessitating manual adjustments.

RT dose optimization

The integration of PET in radiotherapy has led to the development of the concept of biological target volume (BTV), which is based on the biological characteristics of tumors. As BTV can furnish insights into the radiation sensitivity of tumors, it is instrumental in determining the appropriate RT dosage [76]. Utilizing the data derived from BTV, higher RT doses are administered to treatment-resistant regions of the tumor, while lower doses are allocated for treatment-sensitive regions [25, 77]. This heterogeneous RT dose distribution strategy not only ensures that a high local dose is delivered to the tumor but also safeguards vital organs, such as the spinal cord and lungs, consequently mitigating the toxic effects on normal tissues [78].

RT dose calculation

Once the target volume is identified, a dose prescription map becomes essential for dose calculation. Unlike CT, [18F]-FDG PET/MRI lacks the capability to acquire tissue electron density values, dose prescription maps of [18F]-FDG PET/MRI are typically derived through the co-registration of [18F]-FDG PET/MRI and CT images [36, 79]. To streamline the workflow and minimize additional radiation exposure from CT scans, recent advancements suggest the use of pseudo-CT methods generated directly from MR images for dose calculation (Fig. 4) [19, 80, 81]. Ahangari et al. pioneered the transfer of the delineated PTV and optimized treatment plan from CT to pseudo-CT. The dosimetric analysis of the pseudo-CT revealed a mean absolute error within the PTV of 0.17 ± 0.12 Gy, demonstrating a negligible dose difference between the two. This finding was further corroborated by Farjam et al. [19, 82]. In addition, emerging studies indicate that deep learning techniques enable the formulation of pseudo-CT for dose calculation in [18F]-FDG PET/MRI, facilitating the swift acquisition of dose prescription while preserving image quality, though the data is still limited [19, 80, 81, 83, 84].

Fig. 4
figure 4

a The calculated dose plan for CT and pseudo-CT (pCT) images. b Dose and volume histogram (DVH) curve comparing the planned dose of target volumes and different organ at risk (OAR) for CT and pCT

Full size image

Treatment evaluation

In clinical practice, CT and MRI are commonly employed to gauge changes in tumor size post-therapy to evaluate the treatment response. However, as morphological alterations in tumors tend to lag behind hypometabolic changes [85], relying on anatomical data for predicting and assessing treatment response may be inaccuracies. Using specific PET tracers and MRI sequences, [18F]-FDG PET/MRI is capable to capture both tissue morphology and tumor metabolism, making it significant in treatment outcome prediction and evaluation.

Pre-treatment evaluation

The prognosis of advanced cancer patients after treatment is still poor, so it is important to accurately evaluate the treatment effect before treatment, and to guide clinicians in choosing the most appropriate treatment strategies. Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) are robust [18F]-FDG PET/MRI parameters that effectively represent the metabolic activity of tumor cells, providing a comprehensive measure of the overall initial tumor burden before treatment [86]. Meanwhile, as DWI captures the texture of biological tissues [87], the study by Freihat et al. revealed that elevated ADC values derived from DWI, TLG, and MTV before treatment correlated with increased likelihood of recurrence and decreased response rate. This underscores the significance of imaging parameters as valuable predictive biomarkers before treatment, offering more precise information for treatment selection [88]. Recently, image analysis methods such as radiomics and AI have demonstrated their effectiveness in predicting therapy response and prognosis. Features derived from [18F]-FDG PET/MRI radiomics have proven to be viable non-invasive imaging biomarkers for predicting efficacy in cervical cancer treatment. However, additional clinical data are required to validate its feasibility in future applications [89, 90].

Mid-treatment evaluation

To avoid accelerated tumor growth, increased organ toxicity, and increased mortality due to delayed adjustment of therapy [91, 92], it is necessary to evaluate the therapeutic effect and risk of recurrence in the middle of treatment [93, 94]. SUVmax and diffusion-related coefficient, which show glucose metabolism and water diffusion during treatment, can assess tumor sensitivity to the treatment by changes in these parameters. Xu et al. discovered that the minimum diffusion correlation coefficient (Dmin) and SUVmax would be significantly different mid-treatment [86]. In addition, the percentage changes of the maximum standardized uptake value (ΔSUVmax) and the mean diffusion-related coefficient (ΔDmean) of the non-complete metabolic responder during treatment also had a substantial predictive potentiality. Further analysis indicated that the percentage changes in minimum diffusion-related coefficient (ΔDmin) was corrected with ΔSUVmax, suggesting that the two related parameters of [18F]-FDG PET/MRI could be used as substitutes for each other [86]. However, although investigations have shown that [18F]-FDG PET/MRI is useful in treatment evaluation, there is no consensus on the optimal scanning time point [86, 95, 96]. Mayerhoefer et al. demonstrated that capturing treatment-induced changes in lymphoma patients at 48–72 h after the start of treatment could be used to evaluate the therapeutic effect of the first week of treatment [95], while Vojtíšek et al. recorded the pre- and the mid-treatment parameters at week 5 during treatment to predict failure to achieve complete metabolic remission (CMR).

Post-treatment evaluation

The combination of multiple parameters obtained by different image methods is found to be significantly better for evaluating the therapeutic effect [86, 93, 97, 98]. The advent of PET/MRI enables the simultaneous acquisition of multiple parameters while mitigating errors caused by patients’ movement between different scanning modes, thus reducing the impact of physiological changes in tumor tissue due to time intervals [99]. Romeo et al. conducted a study involving patients with head and neck cancer undergoing concurrent chemoradiotherapy. They found that DCE parameters, such as volume transfer constant (Ktrans), and rate constant (Kep), can confirm the treatment response according to morphological, metabolic, and diffusion data. In addition, in combination with elevated post-treatment SUVmax, it can help identify patients with disease progression at follow-up who were initially classified as either in partial response (PR) or stable disease (SD) [100,101,102]. Recently, [18F]-FDG PET/MRI has also been found to be valuable in the evaluation of efficacy after radiotherapy. Mongula et al. found that the use of [18F]-FDG PET/MRI in patients with cervical cancer after radiotherapy could affect the treatment strategy in 50% of patients [103]. Kovács et al. established a mouse model to demonstrate that SUV and ADC values in [18F]-FDG PET/MRI can be used to determine physiological changes in brain tissue after radiotherapy and thus monitor the occurrence of adverse effects after radiotherapy, making [18F]-FDG PET/MRI an irreplaceable tool in the assessment of efficacy after radiotherapy [104]. However, unlike post-chemotherapy efficacy evaluation, radiotherapy may yield up to 50% false positives owing to the occurrence of radiation-induced fibrosis, necrosis, inflammation, and edema [105, 106], potentially limiting the role of [18F]-FDG PET/MRI in assessing radiotherapy efficacy. Nevertheless, recent studies suggest that functional imaging techniques, such as diffusion kurtosis imaging (DKI) and 11C-methionine (11C-MET)-PET, may be beneficial in differentiating residual tumors from radiation-induced fibrotic tissue [103, 107].

In addition, ultrafast DCE-MRI, advanced DWI sequences such as intravoxel incoherent motion (IVIM), diffusion tensor imaging (DTI), proton MR spectroscopy (1H MRS), and chemical exchange saturation transfer (CEST) imaging have shown promise in evaluating the efficacy of cancer patients [108,109,110]. For instance, due to special IVIM parameters, PET/IVIM-MRI, which simultaneously displays the diffusion of water molecules and microcirculation perfusion in the tumor tissue, can be used to evaluate early treatment outcomes for locally advanced cervical cancer and to predict lymphovascular space invasion in cervical cancer without lymphatic metastasis [86, 93, 111]. However, it is noteworthy that the use of these novel imaging sequences in radiotherapy requires further investigation.

Patient surveillance

Patient surveillance plays a key role in the survival of cancer patients. Local recurrence may be the main cause of treatment failure, so timely and accurate restaging of patients with suspected recurrence is required for optimal management. Traditional imaging methods play a decisive role in the evaluation of locoregional recurrence. However, the anatomical changes, scars, and radiation-induced inflammation caused by radiotherapy or surgery make it difficult to distinguish the active tumor tissue, leading to incorrect restaging and thus affecting further treatment decisions [112,113,114]. [18F]-FDG PET/MRI is expected to be the best choice for long-term follow-up of patients. Sawicki et al. showed that in patients with suspected pelvic tumor recurrence, [18F]-FDG PET/MRI correctly diagnosed 100% of malignant lesions, compared with MRI only 74.6%. PET/MRI was also found to reduce the probability of misdiagnosing distant metastases as only local recurrence, thus leading to a higher tumor stage [115]. Since PET/CT was introduced, it has had a high utilization rate in patient follow-up due to its combination of metabolic and anatomical information [116,117,118]. Some studies have compared [18F]-FDG PET/MRI with PET/CT and found that there was no difference in the diagnostic performance of the two in follow-up, but [18F]-FDG PET/MRI can better define the tumor margin, which could show more unclear FDG findings [119]. Additionally, during long-term follow-up, patients may undergo repeated radiation exposure from CT and PET/CT scans, which has been shown to potentially elevate the risk of cancer [120, 121]. Notably, PET/MRI offers a radiation-free alternative, mitigating the potential for cumulative radiation damage and alleviating patient concerns. However, the high purchase and maintenance costs and potential reimbursement issues in PET/MRI could limit its use.

Other works

Specific PET tracers

With the introduction of various tracers, PET/MRI can provide complementary and sensitive information for lesion characterization, boosting its utility in RT planning. As a hypoxic PET tracer, [18F]-Fluoromisonidazole([18F]-FMISO) can predict radiotherapy response, given that normoxic tumors respond better to treatment than hypoxic ones [122]. Neuroendocrine tumors (NETs) frequently overexpress somatostatin receptors (SSTRs). Thus in RT planning of NETs, SSTR-targeted molecular imaging, such as [68Ga]-DOTATOC, [68Ga]-DOTATATE, or [68Ga]-DOTANOC, was used to precisely locate and delineate tumor targets [59]. Recently, the texture features of [68Ga]-DOTATOC-PET/MRI have also been increasingly used to evaluate treatment-related changes in NETs [123]. In addition, as PSMA is highly expressed in prostate cancer (PCa) cells, 68Ga-PSMA-11 PET/MRI provides a good detection rate for PCa in biochemical recurrence after initial curative therapy [124]. Other promising tracers, such as fibroblast activation protein inhibitor-PET/MRI(FAPI-PET/MRI) [125] and 18F-EF5-PET/MRI [126], are currently under investigation but not yet utilized in routine clinical radiotherapy.

Multimodality nanoparticle probes

When the lesion is not well defined from the surrounding tissue, MRI contrast agents are often used to improve image contrast. In integrated PET/MRI, PET tracer and MRI contrast agent are initially used in combination; however, this combination increases both the number of injections and the time spent by patients in the scanner and may ultimately increase costs [127, 128]. To solve these problems, dual-modality probes are under development. Nanoparticles (NPs) lay the foundation for the realization of dual-modality probes due to their highly tunable characteristics and large surface area: volume [129]. In the past few years, iron oxide nanoparticles have been radiolabeled with different radioisotopes (18F, 11C, 13N, 15O, 124I, 64Cu, 68Ga) for cancer diagnosis and evaluation in PET/MRI [130,131,132]. Recently, 68Ga-magnetic iron oxide nanoparticles targeting PSMA and gastrin-releasing peptide receptors (GRPRs) have been developed as a potential tool for PET/MRI diagnosis of PCa and are thought to improve the efficacy of PCa-targeted therapy, but further application in mouse models is needed [128]. In addition, multimodality nanoparticle probes can be concurrently delivered with drugs or therapeutic agents gathering a dual diagnostic and therapeutic effect to perform cancer diagnosis and treatment at the same time [129].

Limitations

Compared with other imaging modalities, although PET/MRI has its outstanding advantages, it still has some inevitable drawbacks. For instance, the inherent drawbacks of PET/MRI make it limiting for some patients. Firstly, a longer image acquisition time makes it less suitable for elderly and pediatric patients who may struggle to remain still for extended periods. Secondly, due to the narrower and longer bore diameter of PET/MRI, some patients who are larger or obese are not candidates for PET/MRI. Thirdly, patients with metal implants, ferromagnetic objects, etc. cannot use PET/MRI, as their presence distorts the magnetic fields and adversely affects image quality. In addition, while the simultaneous collection of PET and MRI data offers potential advantages in motion correction compared to PET/CT, motion artifacts remain an issue. Static motion correction techniques have been suggested as a possible improvement, but these techniques still require further validation [133]. Another notable limitation is the inferiority of MRI compared to CT in the assessment of pulmonary nodules due to the low proton density, magnetic susceptibility, and respiratory motion of lung tissue. Moreover, PET faces difficulties in detecting nodules with low glucose metabolism. Therefore, the precise application of PET/MRI in lung cancer remains an area to be explored [15, 134, 135]. The comparison between PET/MRI and other imaging methods at different angles was shown in Table 3.

Table 3 Comparison of different imaging methods
Full size table

Outlook

Integrated PET/MRI can get both the shape of the tumor and its biological characteristics simultaneously, allowing for a precise understanding of the disease, which in turn guides the implementation of the radiotherapy plan more precisely. With the advent of new technologies, it is anticipated that PET/MRI will be utilized more frequently for radiotherapy.

Availability of data and materials

Not applicable.

Abbreviations

[18F]-FDG:

18F-fluorodeoxyglucose

[18F]-FMISO:

[18F]-Fluoromisonidazole

11C-MET:

11C-methionine

1H MRS:

Proton MR spectroscopy

3D-CRT:

Three-dimensional conformal radiotherapy

AC:

Attenuation correction

ADC:

Apparent diffusion coefficient

AI:

Artificial intelligence

aTL:

Adaptive threshold level

BTV:

Biological target volume

CAD:

Computer-aided design

CEST:

Chemical exchange saturation transfer

CMR:

Complete metabolic remission

CNNs:

Convolutional neural networks

CTV:

Clinical tumor volume

DCE:

Dynamic contrast-enhanced

DKI:

Diffusion kurtosis imaging

Dmin :

Minimum diffusion correlation coefficient

DSC:

Dice similarity coefficient

DTI:

Diffusion tensor imaging

DWI:

Diffusion-weighted imaging

FAPI-PET/MRI:

Fibroblast activation protein inhibitor-PET/MRI

FLAIR:

Fluid-attenuated inversion recovery

FOV:

Field of view

GRPRs:

Gastrin-releasing peptide receptors

GTV:

Gross tumor volume

GTVnd:

Nodal gross tumor volume

IMRT:

Intensity-modulated radiation therapy

IR-GRE:

Inversion recovery gradient echo

IVIM:

Intravoxel incoherent motion

K ep :

Rate constant

K trans :

Volume transfer constant

MRI-UTE:

Magnetic resonance imaging-ultra short echo time

MTV:

Metabolic tumor volume

NETs:

Neuroendocrine tumors

NPs:

Nanoparticles

OAR:

Organs at risk

PCa:

Prostate cancer

PET/MRI:

Positron emission tomography/magnetic resonance imaging

PR:

Partial response

PSMA:

Prostate-specific membrane antigen

PTV:

Planning target volume

RF:

Radiation frequency

RT:

Radiotherapy

s-CT:

Substitute CT

SD:

Stable disease

SI:

Signal strength

SNR:

Signal-to-noise ratio

SRT:

Stereotactic radiotherapy

SSTRs:

Somatostatin receptors

SUV:

Standardized uptake value

SUVmax :

Maximum standardized uptake value

T1WI:

T1-weighted imaging

T2WI:

T2-weighted imaging

TE:

The time of echo

TLG:

Total lesion glycolysis

TSE:

Turbo spin-echo

ΔDmean :

The percentage changes of the mean diffusion-related coefficient

ΔDmin :

The percentage changes in minimum diffusion-related coefficient

ΔSUVmax :

The percentage changes of the maximum standardized uptake value

References

  1. Thariat J, Hannoun-Levi JM, Sun Myint A, Vuong T, Gérard JP (2013) Past, present, and future of radiotherapy for the benefit of patients. Nat Rev Clin Oncol 10:52–60

    Article  CAS  PubMed  Google Scholar 

  2. Schaue D, McBride WH (2015) Opportunities and challenges of radiotherapy for treating cancer. Nat Rev Clin Oncol 12:527–540

    Article  PubMed  PubMed Central  Google Scholar 

  3. Giezen M, Kouwenhoven E, Scholten AN et al (2012) MRI- versus CT-based volume delineation of lumpectomy cavity in supine position in breast-conserving therapy: an exploratory study. Int J Radiat Oncol Biol Phys 82:1332–1340

    Article  PubMed  Google Scholar 

  4. Godinez J, Gombos EC, Chikarmane SA, Griffin GK, Birdwell RL (2008) Breast MRI in the evaluation of eligibility for accelerated partial breast irradiation. AJR Am J Roentgenol 191:272–277

    Article  PubMed  Google Scholar 

  5. Baratto L, Hawk KE, States L et al (2021) PET/MRI improves management of children with cancer. J Nucl Med 62:1334–1340

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Ishiguro T, Nishikawa N, Ishii S et al (2021) PET/MR imaging for the evaluation of cervical cancer during pregnancy. BMC Pregnancy Childbirth 21:288

    Article  PubMed  PubMed Central  Google Scholar 

  7. Cohen D, Kesler M, Muchnik Kurash M, Even-Sapir E, Levine C (2023) A lesson in humility: the added values of PET-MRI over PET-CT in detecting malignant hepatic lesions. Eur J Nucl Med Mol Imaging 50:1423–1433

    Article  PubMed  Google Scholar 

  8. Morawitz J, Bruckmann NM, Dietzel F et al (2022) Comparison of nodal staging between CT, MRI, and [18F]-FDG PET/MRI in patients with newly diagnosed breast cancer. Eur J Nuc Med Mol Imaging 49(3):992–1001

    Article  Google Scholar 

  9. Albano D, Dondi F, Bertagna F, Treglia G (2022) The Role of [68Ga]Ga-Pentixafor PET/CT or PET/MRI in Lymphoma: a systematic review. Cancers 14(15):3814

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Melsaether AN, Raad RA, Pujara AC et al (2016) Comparison of Whole-Body (18)F FDG PET/MR Imaging and Whole-Body (18)F FDG PET/CT in terms of lesion detection and radiation dose in patients with breast cancer. Radiology 281:193–202

    Article  PubMed  Google Scholar 

  11. Botsikas D, Bagetakos I, Picarra M et al (2019) What is the diagnostic performance of 18-FDG-PET/MR compared to PET/CT for the N- and M- staging of breast cancer? Eur Radiol 29:1787–1798

    Article  PubMed  Google Scholar 

  12. Ohno Y, Takeshi Y, Takenaka D, Koyama H, Aoyagi K, Yui M (2020) Comparison of Diagnostic Accuracy for TNM Stage Among Whole-Body MRI and Coregistered PET/MRI Using 1.5-T and 3-T MRI Systems and Integrated PET/CT for Non-Small Cell Lung Cancer. AJR Am J Roentgenol 215:1191–1198

    Article  PubMed  Google Scholar 

  13. Catalano OA, Daye D, Signore A et al (2017) Staging performance of whole-body DWI, PET/CT and PET/MRI in invasive ductal carcinoma of the breast. Int J Oncol 51:281–288

    Article  CAS  PubMed  Google Scholar 

  14. Murthy V, Sonni I, Jariwala N et al (2021) The role of PSMA PET/CT and PET/MRI in the initial staging of prostate cancer. Eur Urol Focus 7:258–266

    Article  PubMed  Google Scholar 

  15. Kim HS, Lee KS, Ohno Y, van Beek EJR, Biederer J (2015) PET/CT versus MRI for diagnosis, staging, and follow-up of lung cancer. J Magn Reson Imaging 42:247–260

    Article  PubMed  Google Scholar 

  16. Fowler AM, Strigel RM (2022) Clinical advances in PET-MRI for breast cancer. Lancet Oncol 23:e32–e43

    Article  PubMed  PubMed Central  Google Scholar 

  17. Winter RM, Leibfarth S, Schmidt H et al (2018) Assessment of image quality of a radiotherapy-specific hardware solution for PET/MRI in head and neck cancer patients. Radiother Oncology 128:485–491

    Article  Google Scholar 

  18. Olin AB, Hansen AE, Rasmussen JH et al (2020) Feasibility of Multiparametric positron emission tomography/magnetic resonance imaging as a one-stop shop for radiation therapy planning for patients with head and neck cancer. Int J Radiat Oncol Biol Phys 108:1329–1338

    Article  PubMed  Google Scholar 

  19. Ahangari S, Hansen NL, Olin AB et al (2021) Toward PET/MRI as one-stop shop for radiotherapy planning in cervical cancer patients. Acta Oncol 60:1045–1053

    Article  CAS  PubMed  Google Scholar 

  20. Thorwarth D, Henke G, Müller A-C et al (2011) Simultaneous 68Ga-DOTATOC-PET/MRI for IMRT treatment planning for meningioma: first experience. Int J Radiat Oncol Biol Phys 81:277–283

    Article  PubMed  Google Scholar 

  21. Paulus DH, Thorwath D, Schmidt H, Quick HH (2014) Towards integration of PET/MR hybrid imaging into radiation therapy treatment planning. Med Phys 41:072505

    Article  PubMed  Google Scholar 

  22. McJury M, O’Neill A, Lawson M et al (2011) Assessing the image quality of pelvic MR images acquired with a flat couch for radiotherapy treatment planning. Br J Radiol 84:750–755

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Blumhagen JO, Braun H, Ladebeck R et al (2014) Field of view extension and truncation correction for MR-based human attenuation correction in simultaneous MR/PET imaging. Med Phys 41:022303

    Article  PubMed  Google Scholar 

  24. Nuyts J, Bal G, Kehren F, Fenchel M, Michel C, Watson C (2013) Completion of a truncated attenuation image from the attenuated PET emission data. IEEE Trans Med Imaging 32:237–246

    Article  PubMed  Google Scholar 

  25. Zamboglou C, Thomann B, Koubar K et al (2018) Focal dose escalation for prostate cancer using Ga-HBED-CC PSMA PET/CT and MRI: a planning study based on histology reference. Radiat Oncol 13:81

    Article  Google Scholar 

  26. Sabater S, Pastor-Juan MR, Berenguer R et al (2019) An MRI comparative image evaluation under diagnostic and radiotherapy planning set-ups using a carbon fibre tabletop for pelvic radiotherapy. Cancer Radiother 23:296–303

    Article  CAS  PubMed  Google Scholar 

  27. Jafar MM, Reeves J, Ruthven MA et al (2016) Assessment of a carbon fibre MRI flatbed insert for radiotherapy treatment planning. Br J Radiol 89:20160108

    Article  PubMed  PubMed Central  Google Scholar 

  28. Eldib M, Bini J, Faul DD, Oesingmann N, Tsoumpas C, Fayad ZA (2016) Attenuation correction for magnetic resonance coils in combined PET/MR imaging: a review. PET Clinics 11:151–160

    Article  PubMed  Google Scholar 

  29. Sander CY, Keil B, Chonde DB, Rosen BR, Catana C, Wald LL (2015) A 31-channel MR brain array coil compatible with positron emission tomography. Magn Reson Med 73:2363–2375

    Article  PubMed  Google Scholar 

  30. Deller TW, Mathew NK, Hurley SA, Bobb CM, McMillan AB (2021) PET image quality improvement for simultaneous PET/MRI with a lightweight MRI surface coil. Radiology 298:166–172

    Article  PubMed  Google Scholar 

  31. Witoszynskyj S, Andrzejewski P, Georg D et al (2019) Attenuation correction of a flat table top for radiation therapy in hybrid PET/MR using CT- and Ge/Ga transmission scan-based μ-maps. Phys Med 65:76–83

    Article  PubMed  Google Scholar 

  32. Wyatt JJ, Howell E, Lohezic M, McCallum HM, Maxwell RJ (2021) Evaluating the image quality of combined positron emission tomography-magnetic resonance images acquired in the pelvic radiotherapy position. Phys Med Biol 66:035018

    Article  PubMed  Google Scholar 

  33. Paulus DH, Oehmigen M, Grüneisen J, Umutlu L, Quick HH (2016) Whole-body hybrid imaging concept for the integration of PET/MR into radiation therapy treatment planning. Phys Med Biol 61:3504–3520

    Article  CAS  PubMed  Google Scholar 

  34. Lebel RM (2020) Performance characterization of a novel deep learning-based MR image reconstruction pipeline. arXiv:2008.06559. https://doi.org/10.48550/arXiv.2008.06559

  35. Wollenweber SD, Delso G, Deller T, Goldhaber D, Hüllner M, Veit-Haibach P (2014) Characterization of the impact to PET quantification and image quality of an anterior array surface coil for PET/MR imaging. MAGMA 27:149–159

    PubMed  Google Scholar 

  36. Alber M, Thorwarth D (2014) Multi-modality functional image guided dose escalation in the presence of uncertainties. Radiother Oncol 111:354–359

    Article  PubMed  Google Scholar 

  37. Martinez-Möller A, Souvatzoglou M, Delso G et al (2009) Tissue classification as a potential approach for attenuation correction in whole-body PET/MRI: evaluation with PET/CT data. J Nucl Med 50:520–526

    Article  PubMed  Google Scholar 

  38. Schulz V, Torres-Espallardo I, Renisch S et al (2011) Automatic, three-segment, MR-based attenuation correction for whole-body PET/MR data. Eur J Nucl Med Mol Imaging 38:138–152

    Article  CAS  PubMed  Google Scholar 

  39. Hofmann M, Steinke F, Scheel V et al (2008) MRI-based attenuation correction for PET/MRI: a novel approach combining pattern recognition and atlas registration. J Nucl Med 49:1875–1883

    Article  PubMed  Google Scholar 

  40. Johansson A, Karlsson M, Nyholm T (2011) CT substitute derived from MRI sequences with ultrashort echo time. Med Phys 38:2708–2714

    Article  PubMed  Google Scholar 

  41. Martinez-Möller A, Souvatzoglou M, Navab N, Schwaiger M, Nekolla SG (2007) Artifacts from misaligned CT in cardiac perfusion PET/CT studies: frequency, effects, and potential solutions. J Nucl Med 48:188–193

    PubMed  Google Scholar 

  42. Beyer T, Antoch G, Blodgett T, Freudenberg LF, Akhurst T, Mueller S (2003) Dual-modality PET/CT imaging: the effect of respiratory motion on combined image quality in clinical oncology. Eur J Nucl Med Mol Imaging 30:588–596

    Article  PubMed  Google Scholar 

  43. Ladefoged CN, Hansen AE, Henriksen OM et al (2020) AI-driven attenuation correction for brain PET/MRI: Clinical evaluation of a dementia cohort and importance of the training group size. Neuroimage 222:117221

    Article  CAS  PubMed  Google Scholar 

  44. Paulus DH, Braun H, Aklan B, Quick HH (2012) Simultaneous PET/MR imaging: MR-based attenuation correction of local radiofrequency surface coils. Med Phys 39:4306–4315

    Article  PubMed  Google Scholar 

  45. Cabello J, Lukas M, Förster S, Pyka T, Nekolla SG, Ziegler SI (2015) MR-based attenuation correction using ultrashort-echo-time pulse sequences in dementia patients. J Nucl Med 56:423–429

    Article  PubMed  Google Scholar 

  46. Kartmann R, Paulus DH, Braun H et al (2013) Integrated PET/MR imaging: automatic attenuation correction of flexible RF coils. Med Phys 40:082301

    Article  PubMed  Google Scholar 

  47. Lindemann ME, Oehmigen M, Lanz T et al (2021) CAD-based hardware attenuation correction in PET/MRI: First methodical investigations and clinical application of a 16-channel RF breast coil. Med Phys 48:6696–6709

    Article  PubMed  Google Scholar 

  48. Li T, Wang J, Yang Y, Glide-Hurst CK, Wen N, Cai J (2023) Multi-parametric MRI for radiotherapy simulation. Med Phys 50:5273–5293

    Article  CAS  PubMed  Google Scholar 

  49. Nevo E, Kamvosoulis P, Currie G (2022) PET/MRI, Part 3: Protocols and Procedures. J Nucl Med Technol 50:17–24

    Article  PubMed  Google Scholar 

  50. Kumar S, Rai R, Stemmer A et al (2017) Feasibility of free breathing Lung MRI for Radiotherapy using non-Cartesian k-space acquisition schemes. Br J Radiol 90:20170037

    Article  PubMed  PubMed Central  Google Scholar 

  51. Ellingson BM, Bendszus M, Boxerman J et al (2015) Consensus recommendations for a standardized Brain Tumor Imaging Protocol in clinical trials. Neuro Oncol 17:1188–1198

    PubMed  PubMed Central  Google Scholar 

  52. Danieli L, Riccitelli GC, Distefano D et al (2019) Brain Tumor-Enhancement Visualization and Morphometric Assessment: a comparison of MPRAGE, SPACE, and VIBE MRI Techniques. AJNR Am J Neuroradiol 40:1140–1148

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Wong KH, Panek R, Bhide SA, Nutting CM, Harrington KJ, Newbold KL (2017) The emerging potential of magnetic resonance imaging in personalizing radiotherapy for head and neck cancer: an oncologist’s perspective. Br J Radiol 90:20160768

    Article  PubMed  PubMed Central  Google Scholar 

  54. Paulson ES, Erickson B, Schultz C, Allen Li X (2015) Comprehensive MRI simulation methodology using a dedicated MRI scanner in radiation oncology for external beam radiation treatment planning. Med Phys 42:28–39

    Article  PubMed  Google Scholar 

  55. Zhang YN, Lu X, Lu ZG, Fu LP, Zhao J, Xiang ZL (2021) Evaluation of Hybrid PET/MRI for Gross Tumor Volume (GTV) Delineation in Colorectal Cancer Liver Metastases Radiotherapy. Cancer Manag Res 13:5383–5389

    Article  PubMed  PubMed Central  Google Scholar 

  56. Zhang S, Xin J, Guo Q, Ma J, Ma Q, Sun H (2014) Comparison of tumor volume between PET and MRI in cervical cancer with hybrid PET/MR. Int J Gynecol Cancer 24:744–750

    Article  PubMed  Google Scholar 

  57. Samołyk-Kogaczewska N, Sierko E, Zuzda K et al (2019) PET/MRI-guided GTV delineation during radiotherapy planning in patients with squamous cell carcinoma of the tongue. Strahlenther Onkol 195:780–791

    Article  Google Scholar 

  58. Ma J-T, Han C-B, Zheng J-H et al (2017) Hybrid PET/MRI-based delineation of gross tumor volume in head and neck cancer and tumor parameter analysis. Nucl Med Commun 38:642–649

    Article  PubMed  Google Scholar 

  59. Mahase SS, Roth O’Brien DA, No D et al (2021) [68Ga]-DOTATATE PET/MRI as an adjunct imaging modality for radiation treatment planning of meningiomas. Neurooncol Adv 3:vdab012

    Article  Google Scholar 

  60. Zhang Y-N, Lu Z-G, Wang S-D et al (2022) Gross tumor volume delineation in primary prostate cancer on 18F-PSMA-1007 PET/MRI and 68Ga-PSMA-11 PET/MRI. Cancer Imaging 22:36

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Afshar-Oromieh A, Wolf MB, Kratochwil C et al (2015) Comparison of 68Ga-DOTATOC-PET/CT and PET/MRI hybrid systems in patients with cranial meningioma: Initial results. Neuro Oncol 17:312–319

    Article  CAS  PubMed  Google Scholar 

  62. Liu L-L, Zhu L-L, Lu Z-G et al (2023) Variability of radiotherapy volume delineation: PSMA PET/MRI and MRI based clinical target volume and lymph node target volume for high-risk prostate cancer. Cancer Imaging 23:1

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Ren J, Eriksen JG, Nijkamp J, Korreman SS (2021) Comparing different CT, PET and MRI multi-modality image combinations for deep learning-based head and neck tumor segmentation. Acta Oncol 60:1399–1406

    Article  CAS  PubMed  Google Scholar 

  64. Song Q, Bai J, Han D et al (2013) Optimal co-segmentation of tumor in PET-CT images with context information. IEEE Trans Med Imaging 32:1685–1697

    Article  PubMed  PubMed Central  Google Scholar 

  65. Bagci U, Udupa JK, Mendhiratta N et al (2013) Joint segmentation of anatomical and functional images: applications in quantification of lesions from PET, PET-CT, MRI-PET, and MRI-PET-CT images. Med Image Anal 17:929–945

    Article  PubMed  PubMed Central  Google Scholar 

  66. Leibfarth S, Eckert F, Welz S et al (2015) Automatic delineation of tumor volumes by co-segmentation of combined PET/MR data. Phys Med Biol 60:5399–5412

    Article  CAS  PubMed  Google Scholar 

  67. Ma DJ, Zhu J-M, Grigsby PW (2011) Tumor volume discrepancies between FDG-PET and MRI for cervical cancer. Radiother Oncol 98:139–142

    Article  PubMed  Google Scholar 

  68. Cavaliere C, Romeo V, Aiello M et al (2017) Multiparametric evaluation by simultaneous PET-MRI examination in patients with histologically proven laryngeal cancer. Eur J Radiol 88:47–55

    Article  PubMed  Google Scholar 

  69. Nestle U, Kremp S, Schaefer-Schuler A et al (2005) Comparison of different methods for delineation of 18F-FDG PET-positive tissue for target volume definition in radiotherapy of patients with non-Small cell lung cancer. J Nucl Med 46:1342–1348

    PubMed  Google Scholar 

  70. Brambilla M, Matheoud R, Secco C, Loi G, Krengli M, Inglese E (2008) Threshold segmentation for PET target volume delineation in radiation treatment planning: the role of target-to-background ratio and target size. Med Phys 35:1207–1213

    Article  CAS  PubMed  Google Scholar 

  71. Kao CH, Hsieh TC, Yu CY et al (2010) 18F-FDG PET/CT-based gross tumor volume definition for radiotherapy in head and neck cancer: a correlation study between suitable uptake value threshold and tumor parameters. Radiat Oncol 5:76

    Article  Google Scholar 

  72. Biehl KJ, Kong FM, Dehdashti F et al (2006) 18F-FDG PET definition of gross tumor volume for radiotherapy of non-small cell lung cancer: is a single standardized uptake value threshold approach appropriate? J Nucl Med 47:1808–1812

    PubMed  Google Scholar 

  73. Alfano R, Bauman GS, Liu W et al (2020) Histologic validation of auto-contoured dominant intraprostatic lesions on [18F] DCFPyL PSMA-PET imaging. Radiother Oncol 152:34–41

    Article  CAS  PubMed  Google Scholar 

  74. Dietlein M, Kobe C, Kuhnert G et al (2015) Comparison of [(18)F]DCFPyL and [(68)Ga]Ga-PSMA-HBED-CC for PSMA-PET imaging in patients with relapsed prostate cancer. Mol Imag Biol 17:575–584

    Article  CAS  Google Scholar 

  75. Sbei A, ElBedoui K, Barhoumi W, Maksud P, Maktouf C (2017) Hybrid PET/MRI co-segmentation based on joint fuzzy connectedness and graph cut. Comput Methods Programs Biomed 149:29–41

    Article  PubMed  Google Scholar 

  76. Pang Y, Wang H, Li H (2021) Medical Imaging Biomarker Discovery and Integration Towards AI-Based Personalized Radiotherapy. Front Oncol 11:764665

    Article  CAS  PubMed  Google Scholar 

  77. Daniel M, Andrzejewski P, Sturdza A et al (2017) Impact of hybrid PET/MR technology on multiparametric imaging and treatment response assessment of cervix cancer. Radiother Oncol 125:420–425

    Article  PubMed  Google Scholar 

  78. van der Heide UA, Houweling AC, Groenendaal G, Beets-Tan RGH, Lambin P (2012) Functional MRI for radiotherapy dose painting. Magn Reson Imaging 30:1216–1223

    Article  PubMed  PubMed Central  Google Scholar 

  79. Leibfarth S, Mönnich D, Welz S et al (2013) A strategy for multimodal deformable image registration to integrate PET/MR into radiotherapy treatment planning. Acta Oncol 52:1353–1359

    Article  PubMed  Google Scholar 

  80. Kemppainen R, Suilamo S, Ranta I et al (2019) Assessment of dosimetric and positioning accuracy of a magnetic resonance imaging-only solution for external beam radiotherapy of pelvic anatomy. Phys Imaging Radiat Oncol 11:1–8

    Article  PubMed  PubMed Central  Google Scholar 

  81. Han X (2017) MR-based synthetic CT generation using a deep convolutional neural network method. Med Phys 44:1408–1419

    Article  CAS  PubMed  Google Scholar 

  82. Farjam R, Tyagi N, Deasy JO, Hunt MA (2019) Dosimetric evaluation of an atlas-based synthetic CT generation approach for MR-only radiotherapy of pelvis anatomy. J Appl Clin Med Phys 20:101–109

    Article  PubMed  Google Scholar 

  83. Edmund JM, Nyholm T (2017) A review of substitute CT generation for MRI-only radiation therapy. Radiat Oncol 12:28

    Article  Google Scholar 

  84. Leynes AP, Yang J, Wiesinger F et al (2018) Zero-Echo-Time and Dixon Deep Pseudo-CT (ZeDD CT): Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI. J Nucl Med 59:852–858

    Article  PubMed  PubMed Central  Google Scholar 

  85. Lee JE, Huh SJ, Nam H, Ju SG (2013) Early response of patients undergoing concurrent chemoradiotherapy for cervical cancer: a comparison of PET/CT and MRI. Ann Nucl Med 27:37–45

    Article  CAS  PubMed  Google Scholar 

  86. Xu C, Sun H, Du S, Xin J (2019) Early treatment response of patients undergoing concurrent chemoradiotherapy for cervical cancer: An evaluation of integrated multi-parameter PET-IVIM MR. Eur J Radiol 117:1–8

    Article  PubMed  Google Scholar 

  87. Queiroz MA, Hüllner M, Kuhn F et al (2014) Use of diffusion-weighted imaging (DWI) in PET/MRI for head and neck cancer evaluation. Eur J Nucl Med Mol Imaging 41:2212–2221

    Article  PubMed  Google Scholar 

  88. Freihat O, Tóth Z, Pintér T et al (2021) Pre-treatment PET/MRI based FDG and DWI imaging parameters for predicting HPV status and tumor response to chemoradiotherapy in primary oropharyngeal squamous cell carcinoma (OPSCC). Oral Oncol 116:105239

    Article  CAS  PubMed  Google Scholar 

  89. Lohmann P, Kocher M, Ruge MI et al (2020) PET/MRI Radiomics in Patients With Brain Metastases. Front Neurol 11:1

    Article  PubMed  PubMed Central  Google Scholar 

  90. Esfahani SA, Torrado-Carvajal A, Amorim BJ et al (2022) PET/MRI and PET/CT Radiomics in Primary Cervical Cancer: A Pilot Study on the Correlation of Pelvic PET, MRI, and CT Derived Image Features. Mol Imag Biol 24:60–69

    Article  CAS  Google Scholar 

  91. Liu FY, Su TP, Wang CC et al (2018) Roles of posttherapy F-FDG PET/CT in patients with advanced squamous cell carcinoma of the uterine cervix receiving concurrent chemoradiotherapy. Eur J Nucl Med Mol Imaging 45:1197–1204

    Article  CAS  PubMed  Google Scholar 

  92. Kirwan JM, Symonds P, Green JA, Tierney J, Collingwood M, Williams CJ (2003) A systematic review of acute and late toxicity of concomitant chemoradiation for cervical cancer. Radiother Oncol 68:217–226

    Article  PubMed  Google Scholar 

  93. Gao S, Du S, Lu Z, Xin J, Gao S, Sun H (2020) Multiparametric PET/MR (PET and MR-IVIM) for the evaluation of early treatment response and prediction of tumor recurrence in patients with locally advanced cervical cancer. Eur Radiol 30:1191–1201

    Article  PubMed  Google Scholar 

  94. Lundemann M, Munck Af Rosenschöld P, Muhic A et al (2019) Feasibility of multi-parametric PET and MRI for prediction of tumour recurrence in patients with glioblastoma. Eur J Nucl Med Mol Imaging 46:603–613

    Article  PubMed  Google Scholar 

  95. Mayerhoefer ME, Raderer M, Jaeger U et al (2018) Ultra-early response assessment in lymphoma treatment: [F]FDG PET/MR captures changes in glucose metabolism and cell density within the first 72 hours of treatment. Eur J Nucl Med Mol Imaging 45:931–940

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  96. Vojtíšek R, Baxa J, Kovářová P et al (2021) Prediction of treatment response in patients with locally advanced cervical cancer using midtreatment PET/MRI during concurrent chemoradiotherapy. Strahlenther Onkol 197:494–504

    Article  Google Scholar 

  97. Herrera FG, Breuneval T, Prior JO, Bourhis J, Ozsahin M (2016) [(18)F]FDG-PET/CT metabolic parameters as useful prognostic factors in cervical cancer patients treated with chemo-radiotherapy. Radiat Oncol 11:43

    Article  Google Scholar 

  98. Liu FY, Lai CH, Yang LY et al (2016) Utility of (18)F-FDG PET/CT in patients with advanced squamous cell carcinoma of the uterine cervix receiving concurrent chemoradiotherapy: a parallel study of a prospective randomized trial. Eur J Nucl Med Mol Imaging 43:1812–1823

    Article  CAS  PubMed  Google Scholar 

  99. Grueneisen J, Beiderwellen K, Heusch P et al (2014) Correlation of standardized uptake value and apparent diffusion coefficient in integrated whole-body PET/MRI of primary and recurrent cervical cancer. PLoS One 9:e96751

    Article  PubMed  PubMed Central  ADS  Google Scholar 

  100. Romeo V, Iorio B, Mesolella M et al (2018) Simultaneous PET/MRI in assessing the response to chemo/radiotherapy in head and neck carcinoma: initial experience. Med Oncol 35:112

    Article  Google Scholar 

  101. Kim S, Loevner LA, Quon H et al (2010) Prediction of response to chemoradiation therapy in squamous cell carcinomas of the head and neck using dynamic contrast-enhanced MR imaging. AJNR Am J Neuroradiol 31:262–268

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  102. King AD, Chow SKK, Yu K-H et al (2015) DCE-MRI for pre-treatment prediction and post-treatment assessment of treatment response in sites of squamous cell carcinoma in the head and neck. PLoS One 10:e0144770

    Article  PubMed  PubMed Central  Google Scholar 

  103. Mongula JE, Bakers FCH, Vöö S et al (2018) Positron emission tomography-magnetic resonance imaging (PET-MRI) for response assessment after radiation therapy of cervical carcinoma: a pilot study. EJNMMI Res 8:1

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  104. Kovács N, Szigeti K, Hegedűs N et al (2018) Multimodal PET/MRI Imaging Results Enable Monitoring the Side Effects of Radiation Therapy. Contrast Media Mol Imaging 2018:5906471

    Article  PubMed  PubMed Central  Google Scholar 

  105. Mongula J, Slangen B, Lambregts D et al (2016) Predictive criteria for MRI-based evaluation of response both during and after radiotherapy for cervical cancer. J Contemp Brachyther 8:181–188

    Article  Google Scholar 

  106. Vincens E, Balleyguier C, Rey A et al (2008) Accuracy of magnetic resonance imaging in predicting residual disease in patients treated for stage IB2/II cervical carcinoma with chemoradiation therapy : correlation of radiologic findings with surgicopathologic results. Cancer 113:2158–2165

    Article  PubMed  Google Scholar 

  107. Dang H, Zhang J, Wang R et al (2022) Glioblastoma Recurrence Versus Radiotherapy Injury: Combined Model of Diffusion Kurtosis Imaging and 11C-MET Using PET/MRI May Increase Accuracy of Differentiation. Clin Nucl Med 47:e428–e436

    Article  PubMed  Google Scholar 

  108. Gollub MJ, Blazic I, Felder S et al (2019) Value of adding dynamic contrast-enhanced MRI visual assessment to conventional MRI and clinical assessment in the diagnosis of complete tumour response to chemoradiotherapy for rectal cancer. Eur Radiol 29:1104–1113

    Article  PubMed  Google Scholar 

  109. Suo S, Cheng F, Cao M et al (2017) Multiparametric diffusion-weighted imaging in breast lesions: association with pathologic diagnosis and prognostic factors. J Magn Reson Imaging 46:740–750

    Article  PubMed  Google Scholar 

  110. Wu B, Warnock G, Zaiss M et al (2016) An overview of CEST MRI for non-MR physicists. EJNMMI Physics 3:19

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  111. Xu C, Yu Y, Li X, Sun H (2021) Value of integrated PET-IVIM MRI in predicting lymphovascular space invasion in cervical cancer without lymphatic metastasis. Eur J Nucl Med Mol Imaging 48:2990–3000

    Article  CAS  PubMed  Google Scholar 

  112. Rangaswamy B, Fardanesh MR, Genden EM et al (2013) Improvement in the detection of locoregional recurrence in head and neck malignancies: F-18 fluorodeoxyglucose-positron emission tomography/computed tomography compared to high-resolution contrast-enhanced computed tomography and endoscopic examination. Laryngoscope 123:2664–2669

    Article  PubMed  Google Scholar 

  113. Sadick M, Schoenberg SO, Hoermann K, Sadick H (2012) Current oncologic concepts and emerging techniques for imaging of head and neck squamous cell cancer. Laryngorhinootologie 91(Suppl 1):S27–S47

    PubMed  Google Scholar 

  114. Herrmann K, Krause BJ, Bundschuh RA, Dechow T, Schwaiger M (2009) Monitoring response to therapeutic interventions in patients with cancer. Semin Nucl Med 39:210–232

    Article  PubMed  Google Scholar 

  115. Sawicki LM, Kirchner J, Grueneisen J et al (2018) Comparison of 18F-FDG PET/MRI and MRI alone for whole-body staging and potential impact on therapeutic management of women with suspected recurrent pelvic cancer: a follow-up study. Eur J Nucl Med Mol Imaging 45:622–629

    Article  CAS  PubMed  Google Scholar 

  116. Cheuk DKL, Sabin ND, Hossain M et al (2012) PET/CT for staging and follow-up of pediatric nasopharyngeal carcinoma. Eur J Nucl Med Mol Imaging 39:1097–1106

    Article  PubMed  PubMed Central  Google Scholar 

  117. Dmytriw AA, Ortega C, Anconina R et al (2022) Nasopharyngeal Carcinoma Radiomic Evaluation with Serial PET/CT: exploring features predictive of survival in patients with long-term follow-up. Cancers 14:3105

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  118. Howard BA, Wong TZ (2021) 18F-FDG-PET/CT Imaging for Gastrointestinal Malignancies. Radiol Clin North Am 59:737–753

    Article  PubMed  Google Scholar 

  119. Queiroz MA, Hüllner M, Kuhn F et al (2014) PET/MRI and PET/CT in follow-up of head and neck cancer patients. Eur J Nucl Med Mol Imaging 41:1066–1075

    PubMed  Google Scholar 

  120. Sodickson A, Baeyens PF, Andriole KP et al (2009) Recurrent CT, cumulative radiation exposure, and associated radiation-induced cancer risks from CT of adults. Radiology 251:175–184

    Article  PubMed  Google Scholar 

  121. Kritsaneepaiboon S, Jutiyon A, Krisanachinda A (2018) Cumulative radiation exposure and estimated lifetime cancer risk in multiple-injury adult patients undergoing repeated or multiple CTs. Eur J Trauma Emerg Surg 44:19–27

    Article  CAS  PubMed  Google Scholar 

  122. Pinker K, Andrzejewski P, Baltzer P et al (2016) Multiparametric [18F]Fluorodeoxyglucose/ [18F]Fluoromisonidazole Positron Emission Tomography/ Magnetic Resonance Imaging of Locally Advanced Cervical Cancer for the Non-Invasive Detection of Tumor Heterogeneity: a pilot study. PLoS One 11:e0155333

    Article  PubMed  PubMed Central  Google Scholar 

  123. Weber M, Kessler L, Schaarschmidt B et al (2020) Treatment-related changes in neuroendocrine tumors as assessed by textural features derived from 68Ga-DOTATOC PET/MRI with simultaneous acquisition of apparent diffusion coefficient. BMC Cancer 20:326

    Article  PubMed  PubMed Central  Google Scholar 

  124. Glemser PA, Rotkopf LT, Ziener CH et al (2022) Hybrid imaging with [68Ga]PSMA-11 PET-CT and PET-MRI in biochemically recurrent prostate cancer. Cancer Imaging 22:53

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  125. Backhaus P, Burg MC, Roll W et al (2022) Simultaneous FAPI PET/MRI Targeting the Fibroblast-Activation Protein for Breast Cancer. Radiology 302:39–47

    Article  PubMed  Google Scholar 

  126. Narva SI, Seppänen MP, Raiko JRH et al (2021) Imaging of Tumor Hypoxia With 18F-EF5 PET/MRI in Cervical Cancer. Clin Nucl Med 46:952–957

    Article  PubMed  Google Scholar 

  127. de Rosales RTM (2014) Potential clinical applications of bimodal PET-MRI or SPECT-MRI agents. J Labelled Compd Radiopharm 57:298–303

    Article  CAS  Google Scholar 

  128. Liolios C, Koutsikou TS, Salvanou EA et al (2022) Synthesis and in vitro proof-of-concept studies on bispecific iron oxide magnetic nanoparticles targeting PSMA and GRP receptors for PET/MR imaging of prostate cancer. Int J Pharm 624:122008

    Article  CAS  PubMed  Google Scholar 

  129. Forte E, Fiorenza D, Torino E et al (2019) Radiolabeled PET/MRI nanoparticles for tumor imaging. J Clin Med 9:89

    Article  PubMed  PubMed Central  Google Scholar 

  130. Karageorgou MA, Vranješ-Djurić S, Radović M et al (2017) Gallium-68 Labeled Iron Oxide Nanoparticles Coated with 2,3-Dicarboxypropane-1,1-diphosphonic Acid as a Potential PET/MR Imaging Agent: a proof-of-concept study. Contrast Media Mol Imaging 2017:6951240

    Article  PubMed  PubMed Central  Google Scholar 

  131. Lee H-Y, Li Z, Chen K et al (2008) PET/MRI dual-modality tumor imaging using arginine-glycine-aspartic (RGD)-conjugated radiolabeled iron oxide nanoparticles. J Nucl Med 49:1371–1379

    Article  CAS  PubMed  Google Scholar 

  132. Bouziotis P, Psimadas D, Tsotakos T, Stamopoulos D, Tsoukalas C (2012) Radiolabeled iron oxide nanoparticles as dual-modality SPECT/MRI and PET/MRI agents. Curr Top Med Chem 12:2694–2702

    Article  CAS  PubMed  Google Scholar 

  133. Martin Vaello R, Reynés-Llompart G, Mateo-Navarro D et al (2018) EP-2204: Impact Of A Motion Correction Method (Q.Static) In The Definition Of The Metabolic Tumor Volume. Radiother Oncol 127:S1217–S1218

    Article  Google Scholar 

  134. Kan X, Ren Y, Li X et al (2022) Value of 18F-FDG PET/MR Imaging in the Early Evaluation of Treatment Response following Radiofrequency Ablation of Liver Cancer in a Rabbit Model. J Vasc Interv Radiol 33:452–459.e3

    Article  PubMed  Google Scholar 

  135. Biederer J, Beer M, Hirsch W et al (2012) MRI of the lung (2/3). Why … when … how? Insights Imaging 3:355–371

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank Dr. Yuan and other colleagues who contributed insightful comments to the paper.

Funding

Not applicable.

Author information

Author notes

  1. Qi Yan and Xia Yan contributed equally to this work.

Authors and Affiliations

  1. Cancer Center, Third Hospital of Shanxi Medical University, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences Tongji Shanxi Hospital, Taiyuan, China

    Qi Yan & Jianbo Song

  2. Cancer Center, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, Taiyuan, China

    Xia Yan & Xin Yang

  3. Shanxi Provincial Key Laboratory for Translational Nuclear Medicine and Precision Protection, Taiyuan, China

    Xia Yan & Jianbo Song

  4. Department of Nuclear Medicine, First Hospital of Shanxi Medical University, Shanxi Medical University, Taiyuan, China

    Sijin Li

Authors
  1. Qi Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xia Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xin Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sijin Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jianbo Song
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

QY and XY jointly contributed to the conceptualization and were the major contributors in writing the manuscript. QY and XY contributed to collecting data. SL and JS contributed to the writing-review. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Sijin Li or Jianbo Song.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

The images we included were approved by the patient.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yan, Q., Yan, X., Yang, X. et al. The use of PET/MRI in radiotherapy. Insights Imaging 15, 63 (2024). https://doi.org/10.1186/s13244-024-01627-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13244-024-01627-6

Keywords