Diffusion-weighted imaging signal as a predictor of pelvic bone marrow activity in middle-aged and elderly patients with rectal cancer

Abstract

Background

Intensity-modulated radiotherapy targeting areas of active bone marrow effectively reduces hematological toxicity; consequently, it is important to determine whether the bone marrow is active.

Purpose

To explore diffusion-weighted imaging (DWI) signal as a potential tool for assessing bone marrow function in middle-aged and elderly patients with rectal cancer.

Material and Methods

A retrospective study investigated clinical and magnetic resonance imaging (MRI) data from middle-aged and elderly patients with rectal cancer. Pelvic bone marrow DWI signals (b = 800 s/mm²) were classified as high and iso-low signal groups. Factors influencing the DWI signal were analyzed individually in a multifactorial analysis. Subsequently, a comparison was made of the intravoxel incoherent motion (IVIM) parameters between the high and iso-low signal groups.

Results

The study involved 73 patients, with 32 in the high-signal and 41 in the iso-low-signal groups. The multifactorial analysis showed that anemia (odds ratio [OR] = 5.264; P = 0.025) and proton density fat fraction (PDFF) (OR = 0.872; P <0.001) were independent factors influencing the DWI signal. In addition, the high-signal group demonstrated significantly lower values of the standard apparent diffusion coefficient (ADC) (median = 0.466 ×10⁻³ mm²/s, interquartile range = 0.413–0.550 vs. 0.534 ×10⁻³ mm²/s, interquartile range = 0.495–0.594; P <0.01) and the mean diffusion coefficient (D) (0.423 ± 0.065 vs. 0.482 ± 0.090, × 10⁻³ mm²/s; P <0.01).

Conclusion

The evaluation of pelvic bone marrow function through DWI signals is feasible in middle-aged and elderly patients with rectal cancer. A high DWI signal in the pelvic bone marrow correlates with post-anemic cellular proliferation, indicating active hematopoiesis.

Keywords

Rectal cancer diffusion-weighted imaging proton density fat fraction anemia active bone marrow

Introduction

Colorectal cancer is the second leading cause of cancer-related death and the third most common malignant tumor worldwide (1). Less than half of cases are locally advanced, requiring combined chemotherapy and radiotherapy (2,3), while some patients benefit from a watchful waiting approach (4). However, radiotherapy-induced hematotoxicity significantly affects colorectal cancer treatment, particularly in rectal cancer (5). This is because the pelvic region contains more than 50% of active bone marrow in adults (6). Research demonstrates a correlation between changes in bone marrow standardized uptake values (SUVs) and post-radiotherapy hematotoxicity, with positron emission tomography-computed tomography (PET-CT)–guided intensity-modulated radiotherapy reducing hematological toxicity by identifying active bone marrow regions (7–11). Despite its effectiveness, PET-CT is costly and associated with risks of radiation exposure (12). Studies have also shown an association between fat content and PET-CT SUVs in bone marrow (13,14). The focus of earlier studies was to identify active bone marrow regions suitable for intensity-modulated radiotherapy using magnetic resonance imaging (MRI)-based fat content (15–17). Evaluating pre-radiotherapy bone marrow fat content is valuable for predicting post-radiotherapy myelosuppression (18,19). In addition, intensity-modulated radiotherapy plans based on fat content to target active bone marrow regions have markedly decreased hematologic toxicity (7,20,21). Nevertheless, recent research suggests that fat content does not reflect differences in bone marrow function between healthy men and women (22). Diffusion-weighted imaging (DWI) signals can offer functional information about bone marrow beyond fat content, with a high DWI signal indicating active hematopoiesis (22–24). The detection of active bone marrow regions using DWI signals holds promise for guiding radiotherapy planning.

Earlier studies have noted high DWI signals in bone marrow, primarily in children and women of childbearing age (25,26). The bone marrow of middle-aged and older individuals typically shows a low DWI signal (25,27). Most rectal cancers typically manifest in middle-aged and older individuals, often diagnosed at locally advanced stages, necessitating radiotherapy. The assessment of bone marrow functional status in those patients is critical for reducing hematological toxicity associated with intensity-modulated radiotherapy. However, the potential for using DWI signals to assess bone marrow function in this particular age group remains unexplored.

Material and Methods

The study was approved by the ethics review board of Jinhua Municipal Central Hospital (no. 20231630101) in accordance with the Declaration of Helsinki. The data used in this study are retrospective and anonymous; therefore, the requirement for informed consent was waived.

Participants

A retrospective analysis was conducted on the data of patients with rectal cancer who underwent rectal MRI between September 2022 and June 2023. The rectal MRI included intravoxel incoherent motion (IVIM) and iterative decomposition of water and fat with echo asymmetry and least squares estimation (IDEAL-IQ) sequences. The exclusion criteria were as follows: patients with rectal cancer after surgery or treatment; patients aged <50 years; patients with bone marrow metastases that were detected on imaging; patients with metallic foreign body or hip replacement artifacts; patients with a history of other malignant tumors or chemoradiotherapy; and those without blood tests in the week before or after the rectal MRI. The flowchart in Fig. 1 shows the initial number of study participants, the excluded participants, and the final study sample.

MRI examination

All participants underwent imaging in a 1.5-T MR scanner (Optima MR360; GE Healthcare, Chicago, IL, USA) using an eight-channel phased-array body coil. Before contrast administration, both IVIM and IDEAL IQ sequences were acquired in the axial plane.

The IVIM sequence was based on standard single-shot spin-echo-planar imaging. Previous studies have concluded that normal bone marrow studies, with relatively low b-values, should be selected (28). The parameters were as follows: b values = 0, 20, 50, 100, 200, 500, 800, and 1000 s/mm²; TR/TE= 5395/76.5 ms; slice thickness = 4 mm; flip angle = 90°; field of view (FOV) = 360 × 360 mm; matrix = 128 × 128; and number of excitations (NEX) = 1, 1, 1, 1, 1, 2, 2, and 4. Fat suppression was achieved during the IVIM sequence using a spectral-spatial excitation pulse.

The IDEAL-IQ parameters were as follows: TR/TE = 15.1/5.9 ms; slice thickness = 5 mm; flip angle = 4°; FOV = 360 × 360 mm; matrix size = 128 × 128; and NEX = 2.

Image evaluation

The GE ADW4.6 workstation was used to determine the maximum dimension of the iliac bone and to define a circular region of interest of approximately 300 ± 50 mm² for IVIM parameters and proton density fat fraction (PDFF) values of both iliac bones.

The IVIM model is as follows:

\frac{s_{b}}{s_{0}} = (1 - f) \times e^{- b \times D} + f \times e^{- b \times D^{*}}

The IVIM model incorporates several parameters, including the standard apparent diffusion coefficient (ADC), which combines both diffusion and perfusion effects. The diffusion coefficient (D) reflects pure molecular diffusion, while the pseudo-diffusion coefficient (D*) represents perfusion-related pseudo-diffusion. In addition, the perfusion fraction (f) indicates the proportion of perfusion (29). Details are shown in Fig. 2.

DWI signals within the pelvic bone marrow at b = 800 s/mm² were classified into four grades based on signal intensity compared to adjacent muscles (25,26): grade 1 = low DWI signal was defined as pelvic marrow signal being lower than that of the neighboring muscles; grade 2 = isosignal indicated pelvic marrow signal being similar to that of adjacent muscles; grade 3 = slightly higher signal indicated pelvic marrow signal slightly higher than that of adjacent muscles; and grade 4 = high signal indicated pelvic bone marrow signal higher than that of surrounding muscles. Grades 1–2 and 3–4 were categorized as the iso-low signal and high signal groups, respectively. The schematic diagram illustrating the classification of DWI signals within the pelvic bone marrow is depicted in Fig. 2.

Observer 1 and observer 2 both have 10 years of experience in MRI diagnosis. For the classification of DWI signals in pelvic bone marrow, their collaborative discussions provided the results. Observer 1 conducted all IVIM and IDEAL-IQ parameters measurements for this study.

Laboratory examination

According to WHO criteria for diagnosing anemia, hemoglobin concentrations <120 g/L in women and <130 g/L in men are defined as anemia (30).

Statistical analysis

Statistical analysis was conducted using SPSS Statistics version 21 (IBM Corp., Armonk, NY, USA). The Shapiro–Wilk test was used to assess the normality of the measures. Age and body mass index (BMI) were compared between high signal and iso-low signal groups using the independent samples t-test. The chi-square test was applied to evaluate the differences in sex, anemic status, T, N, and M stages. Fisher's exact test was employed to evaluate the differences in smoking status. Red cell distribution width (RDW) and PDFF were compared using the Mann–Whitney U-tests. Variables with a P value <0.1 were entered into the multifactor logistic regression analysis using the enter selection method. The differences between D values between high signal and iso-low signal groups were assessed using an independent samples t-test. Differences in standard ADC, D*, and f values were tested using the Mann–Whitney U-tests.

Results

Basic characteristics

The study included a total of 73 patients (49 men, 24 postmenopausal women). Of these cases, 72 were diagnosed with adenocarcinoma and one case was identified as squamous cell carcinoma. The mean age of the patients was 69.00 ± 10.55 years. Anemia was present in 39 patients; of them, 33 diagnosed with mild anemia and six with moderate anemia. The median red blood cell distribution width (RDW) was 13.00 (interquartile range [IQR] = 12.50–14.55).

The DWI signal in the pelvic bone marrow showed a low signal in nine cases, an equal signal in 32 cases, a slightly high signal in 23 cases, and a high signal in nine cases. In this study, 41 (56.16%) cases were categorized in the iso-low signal group, with the remaining 32 (43.84%) cases falling into the high signal group.

Factors affecting the DWI signal in pelvic bone marrow

The univariate analysis showed significant influences on the DWI signal within the pelvic bone marrow by anemia, RDW, and PDFF (all P <0.05). Conversely, age, sex, BMI, smoking status, T, N, and M stages did not show significant effects (all P >0.05). Details are shown in Table 1 and Fig. 3.

Table 1.

The factors influencing DWI signal in pelvic bone marrow.

Characteristic	High signal group (n = 32)	Iso-low signal group (n = 41)	Statistic value	P value
Age (years)	68.50 ± 10.92	69.39 ± 10.38	t = 0.355	0.723
Sex (n, %)			χ² = 0.583	0.445
Men	23 (71.88)	26 (63.41)
Women	9 (28.12)	15 (36.59)
BMI (kg/m²)	23.78 ± 3.55	22.44 ± 3.41	t = −1.635	0.106
Smoking (n, %)			NA	0.689
Yes	2 (6.25)	4 (9.76)
No	30 (93.75)	37 (90.24)
Anemia (n, %)			χ² = 7.795	0.005
Yes	23 (71.88)	16 (39.02)
No	9 (28.12)	25 (60.98)
RDW	13.60 (12.73–15.88)	12.80 (12.50–13.55)	Z = −2.337	0.019
T staging (n, %)			χ² = 0.064	0.801
1–2	7 (21.88)	10 (24.39)
3–4	25 (78.12)	31 (75.61)
N staging (n, %)			χ² = 2.491	0.115
0	12 (37.50)	23 (56.10)
1–2	20 (62.50)	18 (43.90)
M staging			NA	NA
0	32	41	NA	NA
PDFF (%)	60.38 (51.85–68.49)	72.00 (64.00–77.53)	Z = −4.258	< 0.001

Values are given as n (%), mean ± SD, or median (range).

BMI, body mass index; NA, not applicable; PDFF, proton density fat fraction; RDW, red cell distribution width.

The multifactorial analysis revealed that anemia (P = 0.025) and PDFF (P <0.001) were independent risk factors for the DWI signal in pelvic bone marrow, while RDW (P = 0.863) was not. Details are shown in Table 2.

Table 2.

Multifactorial analysis of the factors influencing DWI signal.

Characteristic	OR (95% CI)	P value
Anemia	5.264 (1.228–22.565)	0.025
RDW	1.036 (0.695–1.544)	0.863
PDFF	0.872 (0.808–0.942)	<0.001

CI, confidence interval; DWI, diffusion-weighted imaging; OR, odds ratio; PDFF, proton density fat fraction; RDW, red cell distribution width.

Differences in IVIM parameters between the high and iso-low signal groups

The standard ADC values (P = 0.002) and D values (P = 0.002) were notably lower in the high signal group compared to the iso-low signal group. However, there was no significant variance observed in the D* (P = 0.081) and f (P = 0.677) values between the two groups (Table 3 and Fig. 4).

Table 3.

Differences in IVIM parameters between the high signal and iso-low signal groups.

	High signal group (n = 32)	Iso-low signal group (n = 41)	z/t	P
Standard ADC (10^–3 mm²/s)	0.466 (0.413–0.550)	0.534 (0.495–0.594)	z = −3.041	0.002
D (10^–3 mm²/s)	0.423 ± 0.065	0.482 ± 0.090	t = 3.260	0.002
D* (10^–1 mm²/s)	0.150 (0.132–0.177)	0.168 (0.141–0.214)	z = −1.746	0.081
f	0.280 (0.257–0.320)	0.271 (0.238–0.330)	z = −0.417	0.677

Values are given as mean ± SD or median (range).

ADC, apparent diffusion coefficient; D, diffusion coefficient; D*, pseudodiffusion coefficient; f, perfusion fraction; IVIM, intravoxel incoherent motion.

The clinical and typical images of high signal and low signal in pelvic bone marrow are illustrated in Fig. 5.

Discussion

The present study demonstrated that 44% of middle-aged and elderly patients with rectal cancer exhibited high DWI signals in pelvic bone marrow, and this signal correlates with low PDFF and anemia. This finding aligns with the results of earlier studies (23,24). RDW was significant in univariate analysis but not in multivariate analysis. During the healing process of anemia, smaller red blood cells and prematurely released large reticulocytes from the red bone marrow appear in peripheral blood, resulting in elevated RDW (23). Consequently, RDW is likely a confounding factor related to anemia.

Patients with high DWI signals demonstrated lower ADC and D values. This negative correlation between DWI signal and ADC value is grounded in basic diffusion physics. However, a positive correlation has been reported in healthy women (25–27). First, high DWI signal is common in children and pre-menopausal women and is associated with abundant red marrow and high perfusion (26,31). The presence of T2 shine-through leads to a high DWI signal, while higher perfusion results in increasing ADC values (26). Second, the residual fat signal resulting from non-perfect fat suppression can lead to an underestimation of the ADC value when PDFF increases (32). Consequently, this may lead to a greater underestimation of the ADC values in postmenopausal women. Thus, DWI signals and ADC values exhibited a positive correlation in healthy women. Specifically, our research, which focused on middle-aged and elderly patients, demonstrated relatively low levels of red marrow content and perfusion. It is not surprising to find a negative correlation. The high DWI signal is attributable to restricted water proton mobility resulting from hypercellularity associated with increased hematopoiesis (23), leading to lower ADC values. This aligns with previous studies (23,33). Previous studies have established a positive correlation between f values and the perfusion of bone marrow (34,35). In our study, patients with high DWI signals present with lower ADC and D values, while showing no significant alterations in D* and f values. This high DWI signal may be related to post-anemic cellular proliferation, alongside less significant changes in perfusion, indicating active hematopoiesis.

High DWI signals suggest active bone marrow function in middle-aged and elderly patients with rectal cancer. Evaluating pre-radiotherapy bone marrow function is valuable for predicting post-radiotherapy myelosuppression (18,19). In addition, intensity-modulated radiotherapy plans that target active bone marrow regions have markedly decreased hematologic toxicity (7,20,21). Nevertheless, numerous studies have revealed notable gender-based discrepancies in hematotoxicity among patients with colorectal cancer (5,36). Previous studies assessing bone marrow function through MRI have predominantly focused on fat content (18–21). DWI signal reflects sex-specific functional differences that go beyond fat content in the marrow (22). These findings highlight the importance of using the DWI signal to assess the function of the bone marrow as a complement to the fat content.

The present study has some limitations. First, being a retrospective study, selection bias is inherent in the population selection process. Second, the sample size of this study is relatively small. Moreover, the evaluation of bone marrow function in this study was conducted using IVIM. Although a biopsy may provide a more accurate assessment, it is invasive and not ethical. Finally, as the study focused on rectal cancer, further research is needed to explore its applicability to other pelvic malignancies.

In conclusion, the evaluation of pelvic bone marrow function through DWI signals is feasible in middle-aged and elderly patients with rectal cancer. The high DWI signal in the pelvic bone marrow is associated with lower fat content and anemia status, associated with cell proliferation and implies active hematopoiesis.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication or this article: This research was supported by the Public Service Projects Grant awarded by the Jinhua Science and Technology Bureau (grant nos. 2021-4-033 and 2023-4-092).

ORCID iD

Zheng Han

References

Xie

Chen

Fang

. Comprehensive review of targeted therapy for colorectal cancer. Signal Transduct Target Ther 2020;5:22.

Zygulska

Pierzchalski

. Novel diagnostic biomarkers in colorectal cancer. Int J Mol Sci 2022;23:852.

Kasi

Abbasi

Handa

, et al. Total neoadjuvant therapy vs standard therapy in locally advanced rectal cancer: a systematic review and meta-analysis. JAMA Netw Open 2020;3:e2030097.

Appelt

Pløen

Harling

, et al. High-dose chemoradiotherapy and watchful waiting for distal rectal cancer: a prospective observational study. Lancet Oncol 2015;16:919–927.

Diefenhardt

Martin

Ludmir

, et al. Development and validation of a predictive model for toxicity of neoadjuvant chemoradiotherapy in rectal cancer in the CAO/ARO/AIO-04 phase III trial. Cancers (Basel) 2022;14:4425.

Griffith

. Age-Related changes in the bone marrow. Curr Radiol Rep 2017;5:24.

Miszczyk

Kuna

, et al. Clinical outcomes of pelvic bone marrow sparing radiotherapy for cervical cancer: a systematic review and meta-analysis of randomised controlled trials. Clin Transl Radiat Oncol 2024;47:100801.

Zhou

Zhang

Luo

, et al. Pelvic bone marrow sparing radiotherapy for cervical cancer: a systematic review and meta-analysis. Radiother Oncol 2021;165:103–118.

Beavan

Dundas

Hudson

, et al. Feasibility of bone marrow sparing volumetric modulated arc therapy to spare active bone marrow in cervical and vaginal cancer patients: a retrospective dosimetric analysis. J Med Radiat Sci 2021;68:379–388.

10.

Williamson

Sirák

, et al. Positron emission tomography-guided bone marrow-sparing radiation therapy for locoregionally advanced cervix cancer: final results from the INTERTECC phase II/III trial. Int J Radiat Oncol Biol Phys 2022;112:169–178.

11.

Elicin

Callaway

Prior

, et al. [(18)F]FDG-PET standard uptake value as a metabolic predictor of bone marrow response to radiation: impact on acute and late hematological toxicity in cervical cancer patients treated with chemoradiation therapy. Int J Radiat Oncol Biol Phys 2014;90:1099–1107.

12.

Yusufaly

Miller

Medina-Palomo

, et al. A multi-atlas approach for active bone marrow sparing radiation therapy: implementation in the NRG-GY006 trial. Int J Radiat Oncol Biol Phys 2020;108:1240–1247.

13.

Schraml

Schmid

Gatidis

, et al. Multiparametric analysis of bone marrow in cancer patients using simultaneous PET/MR imaging: correlation of fat fraction, diffusivity, metabolic activity, and anthropometric data. J Magn Reson Imaging 2015;42:1048–1056.

14.

Tsujikawa

Oikawa

Tasaki

, et al. Integrated [(18)F]FDG PET/MRI demonstrates the iron-related bone-marrow physiology. Sci Rep 2020;10:13878.

15.

Lawless

Byrns

Bednarz

, et al. Feasibility of identifying proliferative active bone marrow with fat fraction MRI and multi-energy CT. Phys Med Biol 2024;69. DOI: 10.1088/1361-6560/ad58a0.

16.

Qin

Gong

Wang

, et al. Dosimetric evaluation of bone marrow sparing in proton radiotherapy for cervical cancer guided by MR functional imaging. Radiat Oncol 2022;17:207.

17.

Sajeevan

Singh

Krishnan

, et al. Comparison of CT and MRI for contouring active bone marrow in bone marrow sparing IMRT of carcinoma cervix and its effects on functional outcomes. Eur J Obstet Gynecol Reprod Biol 2022;278:189–194.

18.

Wang

Qin

Gong

, et al. Correlation between changes of pelvic bone marrow fat content and hematological toxicity in concurrent chemoradiotherapy for cervical cancer. Radiat Oncol 2022;17:70.

19.

Gassert

Kranz

Gassert

, et al. Longitudinal MR-based proton-density fat fraction (PDFF) and T2* for the assessment of associations between bone marrow changes and myelotoxic chemotherapy. Eur Radiol 2024;34:2437–2444.

20.

, et al. Effects of bone marrow sparing radiotherapy on acute hematologic toxicity for patients with locoregionally advanced cervical cancer: a prospective phase II randomized controlled study. Radiat Oncol 2024;19:46.

21.

Gao

Wei

Jiang

, et al. Dose-Volume parameters of spared magnetic resonance imaging-defined active bone marrow predict hematologic toxicity in pelvic malignancies patients undergoing radiotherapy: a cohort study. Technol Cancer Res Treat 2024;23:15330338241255283.

22.

Pan

Han

, et al. Impact of fat content on lumbar spine DWI performance: a sex-based comparative study. Eur J Radiol Open 2024;13:100597.

23.

Tsujikawa

Oikawa

Tasaki

, et al. Whole-body bone marrow DWI correlates with age, anemia, and hematopoietic activity. Eur J Radiol 2019;118:223–230.

24.

Chen

Shen

. High signal in bone marrow on diffusion-weighted imaging of female pelvis: correlation with Anemia and fibroid-associated symptoms. J Magn Reson Imaging 2018;48:1024–1033.

25.

Cui

Wang

, et al. Signal characteristics of normal adult bone marrow in whole-body diffusion-weighted imaging. Acta Radiol 2016;57:1230–1237.

26.

Cui

Yao

Cui

, et al. The signal intensity characteristics of normal bone marrow in diffusion weighted imaging at various menstrual status women. Eur J Radiol 2021;143:109938.

27.

Colombo

Bombelli

Summers

, et al. Effects of sex and age on fat fraction, diffusion-weighted image signal intensity and apparent diffusion coefficient in the bone marrow of asymptomatic individuals: a cross-sectional whole-body MRI study. Diagnostics (Basel) 2021;11:913.

28.

Yeung

Wong

Griffith

, et al. Bone marrow diffusion in osteoporosis: evaluation with quantitative MR diffusion imaging. J Magn Reson Imaging 2004;19:222–228.

29.

Le Bihan

. What can we see with IVIM MRI. Neuroimage 2019;187:56–67.

30.

Guideline on haemoglobin cutoffs to define anaemia in individuals and populations. Geneva: World Health Organization, 2024.

31.

Ording Müller

Avenarius

Olsen

. High signal in bone marrow at diffusion-weighted imaging with body background suppression (DWIBS) in healthy children. Pediatr Radiol 2011;41:221–226.

32.

Dieckmeyer

Ruschke

Eggers

, et al. ADC Quantification of the vertebral bone marrow water component: removing the confounding effect of residual fat. Magn Reson Med 2017;78:1432–1441.

33.

Tsujikawa

Makino

Oikawa

, et al. Assessing the ADC of bone-marrow on whole-body MR images in relation to the fat-suppression method and fat content. Magn Reson Med Sci 2022;21:407–413.

34.

Fan

Zhu

Niu

, et al. Correlation of histological marrow characteristics and intravoxel incoherent motion-derived parameters in benign and malignant hematological disorders. Eur J Radiol 2020;123:108745.

35.

Zheng

Niu

, et al. Correlation of intravoxel incoherent motion parameters and histological characteristics from infiltrated marrow in patients with acute leukemia. J Magn Reson Imaging 2020;51:1720–1726.

36.

Dulong

de Souza

Machowiak

, et al. Sex and circadian timing modulate oxaliplatin hematological and hematopoietic toxicities. Pharmaceutics 2022;14:2465.