Skip to main content
Download PDF
Download ePub
  • Research article
  • Open access
  • Published:

Differentiating primary central nervous system lymphoma from glioblastoma by time-dependent diffusion using oscillating gradient

Cancer Imaging volume 23, Article number: 114 (2023) Cite this article

Abstract

Background

This study aimed to elucidate the impact of effective diffusion time setting on apparent diffusion coefficient (ADC)-based differentiation between primary central nervous system lymphomas (PCNSLs) and glioblastomas (GBMs) and to investigate the usage of time-dependent diffusion magnetic resonance imaging (MRI) parameters.

Methods

A retrospective study was conducted involving 21 patients with PCNSLs and 66 patients with GBMs using diffusion weighted imaging (DWI) sequences with oscillating gradient spin-echo (Δeff = 7.1 ms) and conventional pulsed gradient (Δeff = 44.5 ms). In addition to ADC maps at the two diffusion times (ADC7.1 ms and ADC44.5 ms), we generated maps of the ADC changes (cADC) and the relative ADC changes (rcADC) between the two diffusion times. Regions of interest were placed on enhancing regions and non-enhancing peritumoral regions. The mean and the fifth and 95th percentile values of each parameter were compared between PCNSLs and GBMs. The area under the receiver operating characteristic curve (AUC) values were used to compare the discriminating performances among the indices.

Results

In enhancing regions, the mean and fifth and 95th percentile values of ADC44.5 ms and ADC7.1 ms in PCNSLs were significantly lower than those in GBMs (p = 0.02 for 95th percentile of ADC44.5 ms, p = 0.04 for ADC7.1 ms, and p < 0.01 for others). Furthermore, the mean and fifth and 95th percentile values of cADC and rcADC were significantly higher in PCNSLs than in GBMs (each p < 0.01). The AUC of the best-performing index for ADC7.1 ms was significantly lower than that for ADC44.5 ms (p < 0.001). The mean rcADC showed the highest discriminating performance (AUC = 0.920) among all indices. In peritumoral regions, no significant difference in any of the three indices of ADC44.5 ms, ADC7.1 ms, cADC, and rcADC was observed between PCNSLs and GBMs.

Conclusions

Effective diffusion time setting can have a crucial impact on the performance of ADC in differentiating between PCNSLs and GBMs. The time-dependent diffusion MRI parameters may be useful in the differentiation of these lesions.

Background

Glioblastoma (GBM) is the most common and aggressive primary malignant brain tumor in adults [1]. In most cases with clinically and radiographically suspected GBMs, gross surgical resection is attempted. The incidence of primary central nervous system lymphomas (PCNSLs) has significantly increased in both immunosuppressed and immunocompetent individuals, although these lesions are less common than gliomas [2,3,4]. Unlike GBMs, PCNSLs are usually treated with chemotherapy and whole-brain radiotherapy, without extended surgical mass reduction. Therefore, pretreatment differentiation between PCNSLs and GBMs is essential for therapeutic decision-making. On magnetic resonance imaging (MRI), both GBMs and PCNSLs are typically observed as strongly enhanced masses, often accompanied by edema of the surrounding tissue [5,6,7]. Thus, their differentiation may be difficult. Diffusion weighted imaging (DWI) findings and apparent diffusion coefficient (ADC) measurements give valuable information regarding the microstructural organization. The ADC values are known to be inversely correlated with tumor cellularity [8]. Studies have shown that ADC values help differentiate between PCNSLs and GBMs [9,10,11]. PCNSL was characterized by lower ADC than GBM, presumably reflecting higher cellularity [12, 13].

Diffusion time is a basic parameter of DWI and represents the observation time of diffusion. In biological tissues, there are spatial barriers such as fibers and cell membranes that restrict the water molecular motions (restricted diffusion). Under the condition of the restricted diffusion, the ADC values increase when diffusion time decrease [14,15,16,17], as the molecules have less chance to collide with the barriers during the diffusion time. The conventional pulsed gradient spin-echo (PGSE) DWI sequences require a long diffusion time to achieve high b-values due to limited maximum gradient strength [18, 19]. Furthermore, the 180 pulse takes several milliseconds, so there is a limit to the minimum diffusion time in PGSE even if the gradient strength is unlimited. Thus, investigating the effects of diffusion time on ADC using a clinical MRI scanner has been difficult. The oscillating gradient spin-echo (OGSE) DWI sequence is an emerging diffusion encoding method [14], in which quickly oscillating gradients are used instead of the long diffusion-sensitizing gradients used in the PGSE method, thereby allowing for shorter diffusion times. Studies have investigated the diffusion time dependence of ADC using the PGSE and OGSE methods in combination, an approach called time-dependent diffusion MRI. Time-dependent diffusion MRI may add specific information regarding restricted diffusion in the tissue microstructure. To the best of our knowledge, the effect of the diffusion time on the diagnostic performance of ADC in differentiating between PCNSLs and GBMs has not been reported. Moreover, usefulness of the diffusion time dependence of ADC derived from the time dependent MRI for this purpose has never been investigated. Therefore, this study aimed to elucidate the impact of diffusion time setting on ADC-based differentiation between PCNSLs and GBMs and to determine whether time-dependent diffusion MRI is useful in the differentiation of these lesions.

Materials and methods

Patients

We retrospectively evaluated patients with PCNSLs or GBMs who underwent MRI examination between January 2019 and December 2022 at our institution. This retrospective study (approval no. 220126) was approved by our Institutional Review Board, and the need for written informed consent was waived because of the retrospective nature of this study.

Consecutive pathologically proven PCNSLs or GBMs based on the 2021 World Health Organization classification [20] were included in this study. All GBMs were diagnosed based on an integrated diagnosis, combining histology and a glioma-tailored next-generation sequencing panel developed by our institution [21]. The exclusion criteria were as follows: (a) lack of preoperative MRI, including DWI with both OGSE and PGSE sequences; (b) poor image quality; (c) masses smaller than 1 cm; (d) previous surgical resection or irradiation; or (e) lack of contrast-enhancing lesions.

In patients with multiple lesions, the largest mass was examined by MRI.

In this study, 158 consecutive patients (35 with PCNSLs and 123 with GBMs) were considered for inclusion. Seventy-one patients were excluded because of the absence of preoperative MRI, including both OGSE and PGSE DWI scans (11 with PCNSLs and 50 with GBMs), mass smaller than 1 cm (one with PCNSL), poor image quality caused by artifacts in the DWIs (three with GBMs), previous surgical resection or irradiation (two with GBMs), or the lack of contrast-enhancing lesions (two with PCNSLs and two with GBMs). We finally analyzed 21 patients with PCNSLs (15 men and six women; age range 40–87 years; mean age, 70 ± 13 years; 19 with primary diffuse large B-cell lymphoma of the central nervous system, and two with T-cell lymphomas) and 66 patients with isocitrate dehydrogenase-wildtype GMBs (36 men and 30 women; age range 15–92 years; mean age, 70 ± 13 years) (Fig. 1). All patients were diagnosed histopathologically after total or partial surgical resection. All patients were biopsy naive and had not received any treatment before MRI examination. No significant difference in the mean age (p = 0.71) or sex distribution (p = 0.21) was observed between patients with PCNSLs and those with GBMs.

Fig. 1
figure 1

The diagram indicating the inclusion and exclusion criteria and the flow of the inclusion of eligible patients in this study

Full size image

In a previous study, we compared glioblastoma and metastatic brain tumors using the time-dependent diffusion MRI [22]. Patients with glioblastoma analyzed in the present study include 65 patients with glioblastoma used in our previous study.

MRI acquisition

All patients were examined on a 3 T MR scanner (MAGNETOM Prisma; Siemens Healthcare; maximum gradient amplitude = 80 mT/m, maximum slew rate = 200 T/m/s for each gradient axis with a 20-channel head radiofrequency receive coil). DWI was scanned with research sequences for the OGSE DWI using b-values of 0 and 1,500 s/mm2 (number of repeated scans: 1 and 4, respectively) and three diffusion encoding directions. OGSE diffusion encoding used trapezoid-sine waveforms [23]. An effective diffusion time (Δeff) of 7.1 ms (frequency = 50 Hz; diffusion gradient pulse duration [δ] = 8.5 ms) was used. PGSE DWI was also performed with b-values of 0 and 1,500 s/mm2 (number of repeated scans: 1 and 4, respectively) and three diffusion encoding directions. The Δeff for the PGSE encoding was 44.5 ms (diffusion gradient separation [Δ] = 59.8 ms; δ = 46.1 ms). The two sequences used the same parameters, as follows: repetition time (TR), 4,600 ms; echo time (TE), 120 ms; field of view (FOV), 230 × 230 mm2; matrix size, 72 × 72; and slice thickness, 5 mm. The acquisition times were 1 min and 13 s for PGSE DWI and 1 min and 19 s for OGSE DWI. Figure 2 shows the pulse sequence diagrams for PGSE and OGSE.

Fig. 2
figure 2

Schematic representation of the diffusion gradient waveforms for pulsed gradient spin-echo (PGSE) (left) and oscillating gradient spin-echo (OGSE) (right). G = gradient vector; Δ, diffusion gradient separation; δ, diffusion gradient pulse duration; EPI, echo planar imaging; TE, echo time

Full size image

Post-contrast 2D T1-weighted spin-echo images were obtained using the following parameters: TR, 410 ms; TE, 10 ms; number of excitations, 1; matrix, 304 × 304 (reconstructed to 512 × 512); number of slices, 24; slice thickness, 5 mm; interslice gap, 1 mm; FOV, 230 × 230 mm2; and scan time, 2 min and 46 s. These images were used as the anatomical reference in delineating the region of interests (ROIs). Our routine MRI for central nervous system lesions included the following pre-contrast sequences (Table 1): 2D T1-weighted spin-echo imaging, 2D T2-weighted turbo spin-echo imaging, 2D fluid-attenuated inversion recovery (FLAIR) imaging, and 3D susceptibility-weighted imaging. Pre-contrast T1-weighted images were used to confirm contrast enhancement.

Table 1 Imaging parameters of pre- and post-contrast conventional MRI sequences
Full size table

Generating diffusion parametric maps

ADC values were calculated assuming the mono-exponential signal decay between lower and higher b-values.

According to previous studies [24, 25], we evaluated the ADC change (cADC) and the relative ADC change (rcADC) between OGSE (short diffusion time) and PGSE (long diffusion time). cADC and rcADC maps were generated via pixel-by-pixel calculation using the following formulas:

$$\begin{array}{l}\mathrm{cADC}={\mathrm{ADC}}_{7.1\mathrm{ms}}- {\mathrm{ADC}}_{44.5\mathrm{ms}}\\ \begin{array}{cc}\mathrm{rcADC}=\frac{\left({\mathrm{ADC}}_{7.1\mathrm{ms}}- {\mathrm{ADC}}_{44.5\mathrm{ms}}\right)}{ {\mathrm{ADC}}_{44.5\mathrm{ms}}}\times 100 \left(\%\right)\end{array}\end{array}$$

where ADC7.1 ms and ADC44.5 ms are the ADC values obtained using the OGSE and PGSE sequences, respectively.

ROI-based measurement

All images were analyzed using commercially available software (Vitrea; Canon Medical Systems Corporation). The ADC maps were co-registered with the post-contrast T1-weighted images using the rigid body registration. Two independent radiologists (T.H. and Y.K., with 8 and 4 years of radiological experience, respectively), who were blinded to the patients’ clinical and pathological data, performed the ROI analysis. The ROIs were drawn manually on a postcontrast T1-weighted image with the largest tumor diameter, including enhancing region and avoiding necrosis and fluid, and on the corresponding FLAIR image, including non-enhancing peritumoral regions with a FLAIR high signal intensity, and copied them on the corresponding ADC, cADC, and rcADC maps. The mean ADC44.5ms (ADC44.5msmean), ADC7.1ms (ADC7.1msmean), cADC (cADCmean), and rcADC (rcADCmean) were calculated for the entire ROI. Furthermore, the fifth and 95th percentile values of ADC44.5ms (ADC44.5ms5th and ADC44.5ms95th), ADC7.1ms (ADC7.1ms5th and ADC7.1ms95th), cADC (cADC5th and cADC95th), and rcADC (rcADC5th and rcADC95th) were calculated; this method was considered to represent the lowest and highest robust values [26]. The average ROIs size of the enhancing and the peritumoral regions were 587 ± 700 mm2 and 870 ± 668 mm2, respectively, for PCNSLs, and 651 ± 485 mm2 and 611 ± 605 mm2, respectively, for GBMs.

Statistical analysis

The D'Agostino–Pearson normality test was used to check the normality assumption for all parameters in all groups. The Mann–Whitney U test was used to compare the mean age between those with PCNSLs and those with GBMs, and the chi-square test was used to determine sex distribution. The intraclass correlation coefficient (ICC) was used to determine the interobserver agreement between the two observers on parametric measurements. Excellent agreement was defined as ICC > 0.74 [27]. The measurements from the two observers were averaged for each case and were used for further analysis. The ADC values were compared among the different diffusion times using the paired-t test or Wilcoxon signed-rank test. The mean and the 5th and 95th percentile values of ADC44.5 ms, ADC7.1 ms, cADC, and rcADC were compared between PCNSLs and GBMs using the unpaired-t test or Mann–Whitney U test. Receiver operating characteristic (ROC) analysis was performed to determine the optimum threshold for tumor differentiation and to calculate the area under the ROC curve (AUC), sensitivity, specificity, and accuracy for identifying GBM. The best-performing indices were decided for ADC44.5 ms, ADC7.1 ms, cADC, and rcADC. DeLong’s test was used to compare the AUCs of the best-performing indices. The Bonferroni correction was performed to correct multiple comparisons. A commercially available software package (MedCalc, version 15.10.0; MedCalc statistical software) was used for statistical analysis. Differences with p-values of < 0.05 were considered statistically significant.

Results

Figures 3 and 4 show the representative diffusion parametric maps for PCNSLs and GBMs.

Fig. 3
figure 3

A 67-year-old man with primary central nervous system lymphoma. A contrast-enhanced T1-weighted image with a ROI (solid line) (a), a fluid attenuated inversion recovery image with a ROI (dotted line) (b), an ADC map derived from pulsed gradient spin-echo (PGSE) DWI at an effective diffusion time (Δeff) of 44.5 ms (c), an ADC map derived from oscillating gradient spin-echo (OGSE) DWI at an Δeff of 7.1 ms (d), and maps of the ADC change between PGSE DWI and OGSE DWI (cADC) (e) and the relative ADC change between PGSE DWI and OGSE DWI (rcADC) (f). The ADC values in the tumor appear higher at short Δeff values than at long Δeff setting. Large changes in the cADC and rcADC are noted between the OGSE and PGSE sequences in the tumor

Full size image
Fig. 4
figure 4

A 65-year-old woman with glioblastoma, isocitrate dehydrogenase-wildtype, grade 4. A contrast-enhanced T1-weighted image with a ROI (solid line) (a), a fluid attenuated inversion recovery image with a ROI (dotted line) (b), an ADC map derived from pulsed gradient spin-echo (PGSE) DWI at an effective diffusion time (Δeff) of 44.5 ms (c), an ADC map derived from oscillating gradient spin-echo (OGSE) DWI at an Δeff of 7.1 ms (d), and maps of the ADC change between PGSE DWI and OGSE DWI (cADC) (e) and the relative ADC change between PGSE DWI and OGSE DWI (rcADC) (f). The ADC values in the tumor appear higher at short Δeff values than at long Δeff setting. Small changes in the cADC and rcADC are noted between the OGSE and PGSE sequences in the tumor

Full size image

Interobserver agreement

The ICCs and 95% confidence intervals for each parameter are shown in Table 2. All parameters showed an excellent agreement.

Table 2 The intraclass correlation coefficients and 95% confidence intervals for ADC44.5msmean, ADC44.5ms5th, ADC44.5ms95th, ADC7.1msmean, ADC7.1ms5th, ADC7.1ms95th, cADCmean, cADC5th, cADC95th, rcADCmean, rcADC5th, and rcADC95th of the enhancing and peritumoral regions
Full size table

Comparisons of the diffusion parameters

In enhancing regions, all three indices for ADC7.1 ms were significantly higher than those for ADC44.5 ms for both tumors (all p < 0.01) (Fig. 5a–c). All three indices for ADC7.1 ms and ADC44.5 ms were significantly lower for PCNSLs than for GBMs (all p < 0.05) (Fig. 5a–c). Furthermore, all three indices for cADC and rcADC values were significantly higher for PCNSLs than for GBMs (all p < 0.01) (Fig. 5d–i). In peritumoral regions, all three indices for ADC7.1 ms were significantly higher than those for ADC44.5 ms for both tumors (all p < 0.01) (Fig. 6a–c). No significant difference was observed in any of the three indices for ADC44.5 ms, ADC7.1 ms, cADC, and rcADC between PCNSLs and GBMs (Fig. 6a–i).

Fig. 5
figure 5

Box-whisker plots of the ADC44.5msmean and ADC7.1msmean (a), ADC44.5ms5th and ADC7.1ms5th (b), and ADC44.5ms95th and ADC7.1ms95th (c) in enhancing regions for primary central nervous system lymphomas (PCNSLs) and glioblastomas (GBMs). For each tumor, each index for ADC7.1 ms was significantly higher than the corresponding index for ADC44.5 ms (each p < 0.01, respectively) (a-c). Comparisons of the cADCmean (d), cADC5th ©, and cADC95th (f) between PCNSLs and GBMs. Each index for the cADC was significantly higher in PCNSLs than in GBMs (each p < 0.01, respectively). Comparisons of the rcADCmean (g), rcADC5th (h), and rcADC95th (i) for PCNSLs and GBMs. Each index for the rcADC was significantly higher in PCNSLs than in GBMs (each p < 0.01, respectively). Statistical tests used: *paired t-test, †unpaird-t test, ‡Wilcoxon signed-rank test, §Mann–Whitney U test

Full size image
Fig. 6
figure 6

Box-whisker plots of the ADC44.5msmean and ADC7.1msmean (a), ADC44.5ms5th and ADC7.1ms5th (b), and ADC44.5ms95th and ADC7.1ms95th (c) in peritumoral regions for primary central nervous system lymphomas (PCNSLs) and glioblastomas (GBMs). All three indices for ADC7.1 ms were significantly higher than those for ADC44.5 ms for both tumors (all p < 0.01, respectively) (a–c). No significant difference in any of the three indices of ADC44.5 ms, ADC7.1 ms, cADC, and rcADC was observed between PCNSLs and GBMs (a–i). Statistical tests used: *paired t-test, †unpaird-t test, ‡Wilcoxon signed-rank test, §Mann–Whitney U test

Full size image

Diagnostic performance in differentiating PCNSLs from GBMs

Table 3 shows the results of the ROC curve analysis. In enhancing region, the ADC44.5ms5th, ADC7.1ms5th, cADCmean, and rcADCmean values were the best-performing indices for the ADC44.5 ms, ADC7.1 ms, cADC, and rcADC, respectively. Figure 7 shows the ROC curves for the ADC44.5ms5th, ADC7.1ms5th, cADCmean, and rcADCmean values. Pairwise comparisons of the AUCs among the best-performing indices revealed that the AUC of the ADC44.5ms5th was significantly greater than that of the ADC7.1 ms 5th (p < 0.001), whereas no other comparisons of the AUCs revealed significant differences (Table 4). The rcADCmean showed the highest performance (AUC: 0.920).

Table 3 The AUC, optimal threshold, sensitivity, specificity, and accuracy for the ADC44.5msmean, ADC44.5ms5th, ADC44.5ms95th, ADC7.1msmean, ADC7.1ms5th, ADC7.1ms95th, cADCmean, cADC5th, cADC95th, rcADCmean, rcADC5th, and rcADC95th of the enhancing and peritumoral regions to differentiate primary central nervous system lymphomas from glioblastomas
Full size table
Fig. 7
figure 7

Receiver operating characteristic curves for the best-preforming indices for the ADC44.5ms5th, ADC7.1ms5th, cADCmean, and rcADCmean in enhancing regions

Full size image
Table 4 Pairwise comparison of the AUCs among ADC44.5ms5th, ADC7.1ms5th, cADCmean, and rcADCmean of the enhancing regions
Full size table

Discussion

This study showed that the ADC values with both long (44.5 ms) and short (7.1 ms) effective diffusion times (ADC44.5 ms and ADC7.1 ms, respectively) were significantly lower in PCNSLs than in GBMs, and both the cADC and rcADC were significantly higher in PCNSLs than in GBMs. The rcADC, particularly the rcADCmean, showed the highest differentiation performance among all indices, suggesting the clinical usefulness of time-dependent diffusion MRI for this purpose.

The clinical value of ADC in diagnosing PCNSL has been well documented. A distinctly low ADC in the enhancing region of PCNSL reflecting higher cellularity helps differentiate PCNSLs from other brain tumors, including GBMs and metastatic brain tumors [28]. We confirmed that ADCs measured within the enhancing region using the PGSE DWI sequence with an effective diffusion time of 44.5 ms, which is within the range of typical diffusion time of clinical DWI, is useful in distinguishing PCNSLs from GBMs. Notably, our results showed that the diagnostic performance of the ADC7.1 ms obtained using the OGSE DWI sequence was significantly lower than that of the ADC44.5 ms obtained using the conventional PGSE DWI sequence. This finding clearly demonstrates the importance of effective diffusion time setting in clinical DWI. It is presumed that the ADC obtained with shorter effective diffusion times is less sensitive to the microstructural differences between GBMs and PCNSLs.

Previous studies have reported inconsistent results regarding the comparison of ADC in the non-enhancing peritumoral regions between PCNSLs and GBMs. Studies by Ko et al. and Cindil et al. showed that ADC in the peritumoral regions was significantly higher in PCSNLs than in GBMs [29, 30], whereas Wang et al. reported that it did not significantly differ between the two tumor types [31]. The present study failed to show significant difference in any of the peritumoral ADC indices including those of ADC diffusion time dependence between PCNSLs and GBMs. Our results suggest that ADC diffusion time dependence in the peritumoral regions may not be useful for differentiating between PCNSLs and GBMs.

Researchers have reported the usage of time-dependent diffusion MRI in evaluating intracranial tumors. Maekawa et al. showed that both the ADC change and the relative ADC change in ADC between short (6.5 ms) and long (32.5 ms) effective diffusion times were significantly higher in high-grade intra-axial brain tumors than low-grade ones [25]. The same group examined extra-axial tumors using time-dependent diffusion MRI and reported that the relative percentage changes between short (6.5 ms) and long (32.5 ms) effective diffusion times of diffusion tensor eigenvalues (λ1, λ2, and λ3) and the mean diffusivity were significantly higher in pituitary adenomas than in meningothelial meningiomas and acoustic neuromas [32]. More recently, Zhang et al. examined pediatric gliomas using diffusion-time-dependent diffusion MRI and developed a two-compartment microstructural model to obtain intracellular fraction, cell diameter, and cellularity [33]. They demonstrated that the cellularity index achieved the highest performance in identifying the histological grade, and cell diameter achieved the highest discriminating performance for the molecular classification of H3K27-altered gliomas in midline gliomas. Zhu et al. examined five patients with glioma using an ultra-high-performance gradient MRI system, and demonstrated that the ratio of ADC measured at short diffusion times to that measured at long diffusion times is promising for revealing the heterogenous tumor microstructures including cellular density in presurgical and post-treatment gliomas [34]. These studies have suggested the clinical possibility and validity of time-dependent diffusion MRI in characterizing intracranial tumors. However, no studies have investigated the usage of time-dependent diffusion MRI in differentiating PCNSLs from GBMs.

Compared with ADC itself, the diffusion time dependence of ADC is considered more specific to restricted diffusion caused by microscopic barriers in biological tissues, such as cell membranes. Researchers have attributed the stronger diffusion time dependence of ADC in high-grade tumors than in low-grade ones to more abundant diffusion-restricting microstructures within the range of diffusion lengths determined by the selected diffusion times in the OGSE and PGSE DWI sequences [24, 25]. PCNSLs are characterized by monotonous high cellularity and small extracellular space, whereas GBMs are characterized by heterogeneous moderate cellularity and medium-sized extracellular space due to the extracellular matrix, fine hemorrhage, and necrosis [35, 36]. Thus, PCNSLs may have a narrower extracellular space than GBMs, where molecular diffusion is less restricted than that in the intracellular space. The higher cellularity and narrower extracellular space of PCNSLs may account for the stronger diffusion time dependence of ADC. Other reasons for the stronger ADC diffusion time dependence in PCNSLs may include the difference in cell size between PCNSLs and GBMs. Studies have shown that the cell sizes were 10–20 μm for PCNSLs and 10–33 μm for GBMs [37, 38]. The cADC at a given set of diffusion times critically depends on the spacing between the barriers and thus could vary with cell size [39]. At our effective diffusion time settings, the smaller cell size of PCNSLs than GBMs might have resulted in stronger ADC diffusion time dependence. Further studies are needed to discover the pathological basis for our findings.

Limitations of the study

This study had several limitations. First, the sample size was relatively small. Studies with larger sample sizes are required to confirm our findings. Second, only two effective diffusion times (7.1 ms and 44.5 ms) and a fixed set of b-values (0 and 1,500 s/mm2) were investigated. The use of shorter or longer effective diffusion times may have changed the results. However, the gradient performance of our clinical MRI system limits the range of the effective diffusion time in OGSE. Finally, although all tumors were pathologically diagnosed, detailed comparisons between the tissue microstructures and imaging findings were not performed.

Conclusions

The mean of relative changes in the ADC value between short and long diffusion times achieved better performance than ADC from conventional PGSE DWI in differentiating between PCNSLs and GBMs. Our study demonstrated the usefulness of time-dependent diffusion MRI using OGSE in differentiating the two tumor types.

Availability of data and materials

The datasets of current study are available from the corresponding author on reasonable request.

Abbreviations

ADC:

Apparent diffusion coefficient

ADC7.1ms :

Apparent diffusion coefficient values obtained with an effective diffusion time of 7.1 ms

ADC44.5ms :

Apparent diffusion coefficient values obtained with an effective diffusion time of 44.5 ms

cADC:

Apparent diffusion coefficient change

rcADC:

Relative apparent diffusion coefficient change

AUC:

Area under the receiver operating characteristic curve

DWI:

Diffusion weighted imaging

FLAIR:

Fluid attenuated inversion recovery

FOV:

Field of view

GBM:

Glioblastoma

ICC:

Intraclass correlation coefficient

MRI:

Magnetic resonance imaging

OGSE:

Oscillating gradient spin-echo

PGSE:

Pulsed gradient spin-echo

PCNSL:

Primary central nervous system lymphoma

ROC:

Receiver operating characteristic

ROI:

Region of interest

TE:

Echo time

TR:

Repetition time

δ:

Diffusion gradient pulse duration

Δ:

Diffusion gradient separation

Δeff :

Effective diffusion time

References

  1. Thuy MN, Kam JK, Lee GC, Tao PL, Ling DQ, Cheng M, et al. A novel literature-based approach to identify genetic and molecular predictors of survival in glioblastoma multiforme: analysis of 14,678 patients using systematic review and meta-analytical tools. J Clin Neurosci. 2015;22:785–99. https://doi.org/10.1016/j.jocn.2014.10.029.

    Article  CAS  PubMed  Google Scholar 

  2. Cha S, Knopp EA, Johnson G, Wetzel SG, Litt AW, Zagzag D. Intracranial mass lesions: dynamic contrast-enhanced susceptibility weighted echo-planar perfusion MR imaging. Radiology. 2002;223:11–29. https://doi.org/10.1148/radiol.2231010594.

    Article  PubMed  Google Scholar 

  3. Guo AC, Cummings TJ, Dash RC, Provenzale JM. Lymphomas and high-grade astrocytomas: comparison of water diffusibility and histologic characteristics. Radiology. 2002;224:177–83. https://doi.org/10.1148/radiol.2241010637.

    Article  PubMed  Google Scholar 

  4. Morris PG, Abrey LE. Therapeutic challenges in primary CNS lymphoma. Lancet Neurol. 2009;8:581–92. https://doi.org/10.1016/S1474-4422(09)70091-2.

    Article  CAS  PubMed  Google Scholar 

  5. Calli C, Kitis O, Yunten N, Yurtseven T, Islekel S, Akalin T. Perfusion and diffusion MR imaging in enhancing malignant cerebral tumors. Eur J Radiol. 2006;58:394–403. https://doi.org/10.1016/j.ejrad.2005.12.032.

    Article  PubMed  Google Scholar 

  6. Toh CH, Castillo M, Wong AM, Wei KC, Wong HF, Ng SH, et al. Primary cerebral lymphoma and glioblastoma multiforme: differences in diffusion characteristics evaluated with diffusion tensor imaging. AJNR Am J Neuroradiol. 2008;29:471–5. https://doi.org/10.3174/ajnr.A0872.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Yamashita K, Yoshiura T, Hiwatashi A, Togao O, Yoshimoto K, Suzuki SO, et al. Differentiating primary CNS lymphoma from glioblastoma multiforme: assessment using arterial spin labeling, diffusion-weighted imaging, and (1)(8)F-fluorodeoxyglucose positron emission tomography. Neuroradiology. 2013;55:135–43. https://doi.org/10.1007/s00234-012-1089-6.

    Article  PubMed  Google Scholar 

  8. Chen L, Liu M, Bao J, Xia Y, Zhang J, Zhang L, et al. The correlation between apparent diffusion coefficient and tumor cellularity in patients: a meta-analysis. PLoS ONE. 2013;8:e79008. https://doi.org/10.1371/journal.pone.0079008.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Surov A, Ginat DT, Sanverdi E, Lim CCT, Hakyemez B, Yogi A, et al. Use of diffusion weighted imaging in differentiating between malignant and benign meningiomas A multicenter analysis. World Neurosurg. 2016;88:598–602. https://doi.org/10.1016/j.wneu.2015.10.049.

    Article  PubMed  Google Scholar 

  10. Kono K, Inoue Y, Nakayama K, Shakudo M, Morino M, Ohata K, et al. The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol. 2001;22:1081–8.

    CAS  PubMed  PubMed Central  Google Scholar 

  11. Eisenhut F, Schmidt MA, Putz F, Lettmaier S, Fröhlich K, Arinrad S, et al. Classification of primary cerebral lymphoma and glioblastoma featuring dynamic susceptibility contrast and apparent diffusion coefficient. Brain Sci. 2020;10:886. https://doi.org/10.3390/brainsci10110886.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Wen JB, Huang WY, Xu WX, Wu G, Geng DY, Yin B. Differentiating primary central nervous system lymphomas from glioblastomas and inflammatory demyelinating pseudotumor using relative minimum apparent diffusion coefficients. J Comput Assist Tomogr. 2017;41:904–9. https://doi.org/10.1097/RCT.0000000000000636.

    Article  PubMed  Google Scholar 

  13. Doskaliyev A, Yamasaki F, Ohtaki M, Kajiwara Y, Takeshima Y, Watanabe Y, et al. Lymphomas and glioblastomas: differences in the apparent diffusion coefficient evaluated with high b-value diffusion-weighted magnetic resonance imaging at 3T. Eur J Radiol. 2012;81:339–44. https://doi.org/10.1016/j.ejrad.2010.11.005.

    Article  PubMed  Google Scholar 

  14. Baron CA, Beaulieu C. Oscillating gradient spin-echo (OGSE) diffusion tensor imaging of the human brain. Magn Reson Med. 2014;72:726–36. https://doi.org/10.1002/mrm.24987.

    Article  PubMed  Google Scholar 

  15. Portnoy S, Fichtner ND, Dziegielewski C, Stanisz MP, Stanisz GJ. In vitro detection of apoptosis using oscillating and pulsed gradient diffusion magnetic resonance imaging. NMR Biomed. 2014;27:371–80. https://doi.org/10.1002/nbm.3070.

    Article  CAS  PubMed  Google Scholar 

  16. Wu D, Martin LJ, Northington FJ, Zhang J. Oscillating gradient diffusion MRI reveals unique microstructural information in normal and hypoxia-ischemia injured mouse brains. Magn Reson Med. 2014;72:1366–74. https://doi.org/10.1002/mrm.25441.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Pyatigorskaya N, Le Bihan D, Reynaud O, Ciobanu L. Relationship between the diffusion time and the diffusion MRI signal observed at 17.2 Tesla in the healthy rat brain cortex. Magn Reson Med. 2014;72:492–500. https://doi.org/10.1002/mrm.24921.

    Article  PubMed  Google Scholar 

  18. Stejskal EO, Tanner JE. Spin diffusion measurements: spin echoes in the presence of a time-dependent field gradient. J Chem Phys. 1965;42:288–92. https://doi.org/10.1063/1.1695690.

    Article  CAS  Google Scholar 

  19. Stejskal EO. Use of spin echoes in a pulsed magnetic-field gradient to study anisotropic, restricted diffusion and flow. J Chem Phys. 1965;43:3597–603. https://doi.org/10.1063/1.1696526.

    Article  Google Scholar 

  20. Cree IA, Lokuhetty D, Peferoen LAN, et al. WHO classification of tumours of the central nervous system. 5th ed. Lyon: IARC 2021:15–187, 349–69.

  21. Higa N, Akahane T, Yokoyama S, Yonezawa H, Uchida H, Takajo T, et al. A tailored next-generation sequencing panel identified distinct subtypes of wildtype IDH and TERT promoter glioblastomas. Cancer Sci. 2020;111:3902–11. https://doi.org/10.1111/cas.14597.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Kamimura K, Kamimura Y, Nakano T, Hasegawa T, Nakajo M, Yamada C, et al. Differentiating brain metastasis from glioblastoma by time-dependent diffusion MRI. Cancer Imaging. 2023;23:75. https://doi.org/10.1186/s40644-023-00595-2.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Does MD, Parsons EC, Gore JC. Oscillating gradient measurements of water diffusion in normal and globally ischemic rat brain. Magn Reson Med. 2003;49:206–15. https://doi.org/10.1002/mrm.10385.

    Article  PubMed  Google Scholar 

  24. Iima M, Yamamoto A, Kataoka M, Yamada Y, Omori K, Feiweier T, et al. Time-dependent diffusion MRI to distinguish malignant from benign head and neck tumors. J Magn Reson Imaging. 2019;50:88–95. https://doi.org/10.1002/jmri.26578.

    Article  PubMed  Google Scholar 

  25. Maekawa T, Hori M, Murata K, Feiweier T, Kamiya K, Andica C, et al. Differentiation of high-grade and low-grade intra-axial brain tumors by time-dependent diffusion MRI. Magn Reson Imaging. 2020;72:34–41. https://doi.org/10.1016/j.mri.2020.06.018.

    Article  PubMed  Google Scholar 

  26. Chung WJ, Kim HS, Kim N, Choi CG, Kim SJ. Recurrent glioblastoma: optimum area under the curve method derived from dynamic contrast-enhanced T1-weighted perfusion MR imaging. Radiology. 2013;269:561–8. https://doi.org/10.1148/radiol.13130016.

    Article  PubMed  Google Scholar 

  27. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420–8. https://doi.org/10.1037//0033-2909.86.2.420.

    Article  CAS  PubMed  Google Scholar 

  28. Shim WH, Kim HS, Choi CG, Kim SJ. Comparison of apparent diffusion coefficient and intravoxel incoherent motion for differentiating among glioblastoma, metastasis, and lymphoma focusing on diffusion-related parameter. PLoS ONE. 2015;30:e0134761. https://doi.org/10.1371/journal.pone.0134761.

    Article  CAS  Google Scholar 

  29. Ko CC, Tai MH, Li CF, Chen TY, Chen JH, Shu G, et al. Differentiation between Glioblastoma Multiforme and Primary Cerebral Lymphoma: Additional Benefits of Quantitative Diffusion-Weighted MR Imaging. PLoS ONE. 2016;11:e0162565. https://doi.org/10.1371/journal.pone.0162565.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Cindil E, Sendur HN, Cerit MN, Dag N, Erdogan N, Celebi FE, et al. Validation of combined use of DWI and percentage signal recovery-optimized protocol of DSC-MRI in differentiation of high-grade glioma, metastasis, and lymphoma. Neuroradiology. 2021;63:331–42. https://doi.org/10.1007/s00234-020-02522-9.

    Article  PubMed  Google Scholar 

  31. Wang S, Kim S, Chawla S, Wolf RL, Knipp DE, Vossough A, et al. Differentiation between glioblastomas, solitary brain metastases, and primary cerebral lymphomas using diffusion tensor and dynamic susceptibility contrast-enhanced MR imaging. AJNR Am J Neuroradiol. 2011;32:507–14. https://doi.org/10.3174/ajnr.A2333.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Maekawa T, Hori M, Murata K, Feiweier T, Kamiya K, Andica C, et al. Investigation of time-dependent diffusion in extra-axial brain tumors using oscillating-gradient spin-echo. Magn Reson Imaging. 2023;96:67–74. https://doi.org/10.1016/j.mri.2022.11.010.

    Article  PubMed  Google Scholar 

  33. Zhang H, Liu K, Ba R, Zhang Z, Zhang Y, Chen Y, et al. Histological and molecular classifications of pediatric glioma with time-dependent diffusion MRI based microstructural mapping. Neuro Oncol. 2023;7:noad003. https://doi.org/10.1093/neuonc/noad003.

    Article  Google Scholar 

  34. Zhu A, Shih R, Huang RY, DeMarco JK, Bhushan C, Morris HD, et al. Revealing tumor microstructure with oscillating diffusion encoding MRI in pre-surgical and post-treatment glioma patients. Magn Reson Med. 2023;90:1789–801. https://doi.org/10.1002/mrm.29758.

    Article  CAS  PubMed  Google Scholar 

  35. Makino K, Hirai T, Nakamura H, Kuroda JI, Shinojima N, Uetani H, et al. Differentiating between primary central nervous system lymphomas and glioblastomas: combined use of perfusion-weighted and diffusion-weighted magnetic resonance imaging. World Neurosurg. 2018;112:e1–6. https://doi.org/10.1016/j.wneu.2017.10.141.

    Article  PubMed  Google Scholar 

  36. Cha YJ, Choi J, Kim SH. Presence of apoptosis distinguishes primary central nervous system lymphoma from glioblastoma during intraoperative consultation. Clin Neuropathol. 2018;37:105–11. https://doi.org/10.5414/NP301075.

    Article  PubMed  Google Scholar 

  37. Sato Y, Ura S, Nishi K. Textbook of cytopathology. 4th ed. Tokyo: Kindai syuppan; 2018.

    Google Scholar 

  38. Inagaki T, Ohtsu K, Hara M, Shimoyama K, Harada A, Shimizu K, et al. A proposed study of normal neurocytology of the central nervous system, and comparison to the cytology of astrocytomas. J Jpn Soc Clin Cytol. 1996;5:105–13. https://doi.org/10.5795/jjscc.35.105.

    Article  Google Scholar 

  39. Xu J, Does MD, Gore JC. Dependence of temporal diffusion spectra on microstructural properties of biological tissues. Magn Reson Imaging. 2011;9:380–90. https://doi.org/10.1016/j.mri.2010.10.002.

    Article  Google Scholar 

Download references

Acknowledgements

The authors wish to thank the staffs of Kagoshima University Hospital for their support.

Funding

Not applicable.

Author information

Authors and Affiliations

  1. Department of Advanced Radiological Imaging, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan

    Kiyohisa Kamimura & Takashi Yoshiura

  2. Department of Radiology, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan

    Tsubasa Nakano, Tomohito Hasegawa, Masanori Nakajo, Chihiro Yamada, Yoshiki Kamimura, Kentaro Akune, Fumitaka Ejima, Takuro Ayukawa, Hiroaki Nagano, Koji Takumi, Masatoyo Nakajo & Takashi Yoshiura

  3. Department of Neurosurgery, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan

    Nayuta Higa, Hajime Yonezawa & Ryosuke Hanaya

  4. Department of Pathology, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan

    Mari Kirishima & Akihide Tanimoto

  5. Department of Radiological Technology, Kagoshima University Hospital, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan

    Takashi Iwanaga

  6. Siemens Healthcare K.K., Gate City Osaki West Tower, 1-11-1 Osaki, Shinagawa-Ku, Tokyo, 141-8644, Japan

    Hiroshi Imai

  7. Siemens Healthcare GmbH, Henkestrasse 127, 91052, Erlangen, Germany

    Thorsten Feiweier

Authors
  1. Kiyohisa Kamimura
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Tsubasa Nakano
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Tomohito Hasegawa
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Masanori Nakajo
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Chihiro Yamada
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yoshiki Kamimura
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Kentaro Akune
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Fumitaka Ejima
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Takuro Ayukawa
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Hiroaki Nagano
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Koji Takumi
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Masatoyo Nakajo
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Nayuta Higa
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Hajime Yonezawa
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Ryosuke Hanaya
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Mari Kirishima
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Akihide Tanimoto
    View author publications

    You can also search for this author in PubMed Google Scholar

  18. Takashi Iwanaga
    View author publications

    You can also search for this author in PubMed Google Scholar

  19. Hiroshi Imai
    View author publications

    You can also search for this author in PubMed Google Scholar

  20. Thorsten Feiweier
    View author publications

    You can also search for this author in PubMed Google Scholar

  21. Takashi Yoshiura
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Kiyohisa Kamimura: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft; Tsubasa Nakano: Data curation; Tomohito Hasegawa: Data curation; Masanori Nakajo: Data curation; Chihiro Yamada: Data curation; Yoshiki Kamimura: Data curation; Kentaro Akune: Data curation; Fumitaka Ejima: Data curation; Takuro Ayukawa: Data curation; Hiroaki Nagano: Data curation; Koji Takumi: Data curation; Masatoyo Nakajo: Data curation; Hiroyuki Uchida: Data curation; Nayuta Higa: Data curation; Hajime Yonezawa: Data curation; Ryosuke Hanaya: Data curation; Mari Kirishima: Data curation; Akihide Tanimoto: Data curation; Takashi Iwanaga: Data curation; Hiroshi Imai: Software, Visualization; Thorsten Feiweier: Software, Visualization; Takashi Yoshiura: Conceptualization, Funding acquisition, Methodology, Project administration, Writing – review & editing. The authors read and approved the final manuscript.

Corresponding author

Correspondence to Kiyohisa Kamimura.

Ethics declarations

Ethics approval and consent to participate

This retrospective study was approved by our institutional ethics review board (Ethics Committee on Epidemiological Studies Kagoshima University Graduate School of Medical and Dental Sciences, No. 220126), and the requirement for informed consent of patients was waived.

Consent for publication

Not applicable.

Competing interests

Hirosi Imai is an employee of Siemens Healthcare K.K. and Thorsten Feiweier is an employee of Siemens Healthcare GmbH, owns stocks of Siemens (Healthineers) AG and holds patents filed by Siemens.

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/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kamimura, K., Nakano, T., Hasegawa, T. et al. Differentiating primary central nervous system lymphoma from glioblastoma by time-dependent diffusion using oscillating gradient. Cancer Imaging 23, 114 (2023). https://doi.org/10.1186/s40644-023-00639-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40644-023-00639-7

Keywords

Cancer Imaging

ISSN: 1470-7330

Contact us