The value of magnetic resonance blood oxygen level-dependent imaging in evaluating the efficacy of advanced cervical cancer combined with radiotherapy and chemotherapy

Abstract

Background

Blood oxygen level-dependent magnetic resonance imaging (BOLD-MRI) is an imaging method used to analyze oxygenation status of the tumor.

Purpose

To investigate the feasibility of BOLD-MRI in evaluating the efficacy of advanced cervical cancer combined with radiotherapy and chemotherapy.

Material and Methods

This prospective study included 85 patients with advanced cervical cancer who received BOLD-MRI examination before and after concurrent chemoradiotherapy from October 2020 to December 2021. To investigate the changes of baseline R2* values and △R2* values of cervical cancers before and after treatment.

Results

29 cases were complete response, 34 cases were partial response, and 22 cases showed progression. The baseline R2* values of the tumors were lower than that of the normal cervical muscle (P < 0.0001). After oxygen stimulation, the baseline R2* values of the tumors decreased (P = 0.012). After treatment, the baseline R2* values of the tumors increased (P = 0.007), and the dynamic △R2* values of the tumors decreased (P = 0.025). The baseline R2* value of the complete response was the highest (P = 0.000), the dynamic △R2* value of the complete response was the lowest (P = 0.017).

Conclusion

BOLD-MRI can evaluate the efficacy of concurrent chemoradiotherapy for advanced cervical cancer.

Keywords

Magnetic resonance blood oxygen level-dependent imaging hypoxia function parameters cervical cancer radiotherapy and chemotherapy efficacy evaluation

Introduction

Cervical cancer is a commonly diagnosed tumor of the female reproductive system. According to 2018 global cancer statistics, its incidence rate and mortality rank fourth among female malignant tumors (1). Tumor hypoxia is a common phenomenon of solid tumors, which is of significance for tumor metastasis, curative effect, and prognosis (2,3). Its occurrence is related to the degree, grade, and stage of tumor malignancy (4). Hypoxic metabolism is one of the main causes of recurrence, metastasis, and chemoradiotherapy resistance in cervical cancer (5 –7). Therefore, imaging studies that are based on hypoxic metabolism are of considerable value for the treatment and prognosis of patients with cervical cancer.

Blood oxygen level-dependent magnetic resonance imaging (BOLD-MRI) is an imaging method used to analyze vasculature oxygenation concentration information (8 –10). BOLD imaging can sensitively detect the change in blood volume, blood flow, and blood oxygen using deoxyhemoglobin as an internal paramagnetic contrast agent (9–10). The increase of deoxyhemoglobin concentration in the blood accelerates its transverse relaxation rate (T2*). The change in the ratio of deoxyhemoglobin concentration in tumor vessels (R2*) reflects the oxygen supply of the tumor (10). Therefore, the T2* or R2* value is related to the concentration of deoxyhemoglobin in tumor blood vessels, which can indirectly reflect the oxygen distribution state of tumor tissue (11). Studies have confirmed the ability of BOLD imaging to non-invasively and quantitatively evaluate the overall hypoxia of a tumor (9 –12). Because BOLD imaging can reflect the oxidation of the lesion itself, it can be used to evaluate the prognosis of cervical cancer, but large-scale clinical trials are still needed to clarify its specific role (5 –7).

The aim of the present study was to explore the feasibility of BOLD imaging as a non-invasive method for evaluating the hypoxic state of advanced cervical cancer in a large sample and further determine the value of the R2* quantitative parameter in predicting the efficacy of concurrent radiotherapy and chemotherapy for patients with cervical cancer.

Material and Methods

Research objective

This prospective study included patients with cervical cancer who were treated in the authors’ hospital from October 2020 to December 2021. This study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of The First Hospital of Lanzhou University, and informed consent was obtained from all participants.

The study's inclusion criteria were as follows: (i) patients who had been diagnosed as having stage IIB cancer or above, as defined by the International Federation of Gynecology and Obstetrics (FIGO); (ii) patients who received radical concurrent chemoradiotherapy; and (iii) patients who had been fully informed of the study content and provided signed informed consent voluntarily for inclusion in the research.

The study's exclusion criteria were as follows: patients with contraindications for MR examination were excluded.

An MR examination was performed for all the patients included in the study before radical concurrent chemoradiotherapy and again two months after the start of treatment.

Magnetic resonance scanning protocol

All patients received a GE Signa Architect 3.0-T high field MR scan (GE Healthcare, Florence, South Carolina, USA). The body coil was used as the radio frequency transmitting coil, and the abdominal and pelvic multi-channel phased array 16-channel coil was used as the receiving coil. Check for breathing exercises before the scan to minimize abdominal movements during the scan. 1. Scan range: from pubic symphysis to para-aorta at the level of renal hilum. Axial T1WI: TR = 550ms, TE =13ms, layer thickness = 4mm, spacing =1mm, field of view（FOV）= 400mm×400mm, matrix = 195×320, excitation times=2. Fat suppression sequence in axial and sagittal T2WI of organs: TR = 550ms, TE = 3ms, layer thickness=1mm, spacing =1mm, field of view (FOV) = 400mm×400mm, matrix = 202×384, excitation times=2. Diffusion weighted imaging: single-shot axial excitation SE-EPI sequence, TR = 3500ms, TE = 93ms, layer thickness =4mm, spacing = 1mm, field of view (FOV) = 400 mm × 400 mm, matrix = 320 × 256, three directions of diffusion sensitive gradients were selected. b values are 0, 800 and 1000s/mm², respectively.

The BOLD scanning parameters in this study were as follows: R2* (T2*W) multi-TE = 2, 4, 6, 8, 10, 11, 13, and 15 ms; TR = 43.6 ms; layer thickness = 4 mm; layer spacing = 1 mm; layer number = 17; FOV = 380 × 380 mm; matrix = 256 × 192; the number of incentives = 1; flip angle = 25°; scanning time = 1 min and 50 s. The scanning range was centered on the maximum section of the tumor.

The blood oxygen level-dependent imaging process

Before the examination, the patient was fully informed about the process and any discomfort they may experience. Before entering the magnet, the patient was equipped with a positive-pressure oxygen mask. The mask was gently but snugly attached to the face. A rubber hose was connected to a oxygen cylinder outside the examination room. The cylinder was equipped with a specific pressure valve and a measuring instrument to control the oxygen flow rate and flow rate.

First, BOLD image scanning was performed under normal conditions (breathing air), equivalent to a plain scan under normal conditions. Then, the valve was opened to allow patients to inhale the high concentration of oxygen (concentration >95%, Lanzhou Fangyuan Medical Special Gas Company, Lanzhou, China), at a flow rate of 10 L/min. During the inspiratory process, researcher remained in the room with patients to ease any tension caused by excessive ventilation.

Pipeline loss was considered, and oxygen inhalation time was set at 10 min. After the inhalation was completed, the patient underwent BOLD image scanning for a second time.

Image postprocessing

Using the R2 Star software of the GE AW VolumeShare 7 workstation, the images and data were processed, measured, and analyzed using the same method of negotiation by two radiologists with more than 10 years of experience in gynecological MRI. The mean R2* values obtained by the two radiologists were used for statistical analysis. The R2* value of the tumor hypoxia function parameter was calculated and generated by manually selecting appropriate regions of interest (ROIs). According to T2-weighted fat suppression imaging, the R2* values of tumor (baseline R2* values) and normal uterine muscle during air inhalation were measured on the color-coded R2* maps. After the effects of oxygen stimulation, the R2* values of tumor were measured and the dynamic ΔR2* values of oxygen stimulation were calculated. Namely, ΔR2* value = R2* value before inhalation-R2* value after inhalation. The solid part of the tumor was measured, and 2–3 ROIs were selected according to the tumor size at various levels. Each ROI was placed in a different tumor region with intervals of ≥5 mm, covering as much of the tumor as possible while avoiding tumor margins, necrosis, bleeding, visible blood vessels, or uterine cavity. Measurements taken after treatment were the same as before treatment. If the tumor had completely retracted after treatment, the ROI was collected in the corresponding region in the control site of the tumor before treatment. On color-coded R2* maps, the red areas indicate the highest R2* value, reflecting a high concentration of deoxyhemoglobin, while blue areas indicate the lowest R2* value, reflecting a low concentration of deoxyhemoglobin. To ensure the accuracy of measurements, the area range of each ROI was 11–14 mm²_.

Radiochemotherapy

Radiotherapy was conducted as follows.

Conformal radiotherapy: In vitro three dimensional intensity-modulated conformal radiotherapy combined with afterloading was used. Intensity-modulated radiotherapy protocol: the total dose of 45-50Gy, 1.8-2Gy each time, 5 times a week, for 5-6 weeks, with additional dose of 10-20 Gy if necessary. Post-loading internal radiotherapy protocol: high-dose-rate brachytherapy was used once A week, with a point A dose of 6Gy each time for 5 weeks, with a total dose of 30Gy. The total dose of external and endovascular radiotherapy was not less than 75Gy. Chemotherapy regimen: cisplatin 40mg/m², once a week, for 6 weeks.

Evaluation of clinical efficacy: Referring to follow-up records exceeding 6 months, a gynecologist and a radiologists evaluated the clinical efficacy together. Radiological examinations, clinical symptoms, Physical examinations, Pap smears, and serum tumor markers were also comprehensively evaluated. Patients were classified into different groups (13): complete response(completely remission without recurrence or metastasis), partial response(partial remission without metastasis), and progression(local-regional progressed or metastasis).

Statistical methods

The SPSS Statistics version 26.0 software program (IBM Corp., Armonk, NY, USA) was used to conduct the statistical analysis. An independent sample t-test was used to compare the baseline R2* values of tumor tissue and uterine muscular tissue before treatment. Paired t-tests were used to compare the changes in R2* values before and after oxygen stimulation before treatment, as well as changes in the baseline R2* values and dynamic △R2* values of tumor tissue before and after treatment. Two groups of patients with and without responde were compared to basic informations using the t test, Fisher's exact test, Wilcoxon rank sum test, or chi-square tests. P < 0.05 indicated a statistically significant difference.

Results

Basic patient information

A total of 85 patients (age range = 37–63 years; mean age = 48.7 ± 11.2 years) with cervical cancer were included in this study, including 26 cases of stage IIb, 19 cases of stage IIIa, 14 cases of stage IIIb, 8 cases of stage IIIc, 11 cases of stage IVa, and 7 cases of IVb as defined (FIGO). There were 21 postmenopausal patients. After concurrent chemoradiotherapy and 6 months of follow-up, 29 cases were considered to complete response, 34 cases were partial response, and 22 cases showed progression. The patients were divided into two groups the responders , which combined the complete response and partial response, and the non-responders comprising the tumor progression. According to the results in Table 1, the patient's age, systolic blood pressure, FIGO stage, tumor volume, and whether or not menopause had an impact on the efficacy.

Table 1.

Basic information of two groups of patients.

	Responders（n=63）	Non-responders（n=22）	t/x²	p
Age	39.87±4.97	48.87±8.27	5.887	0.000*
IBM	23.97±4.20	22.43±4.33	1.081	0.377
blood pressure
systolic pressure	123.17±9.10	131.08±11.09	3.861	0.003*
diastolic pressure	81.11±5.20	86.33±8.84	1.164	0.091
Pathological classification			1.288	0.525
Squamous cell carcinoma	46	14
Adenocarcinoma	13	5
Adenosquamous carcinoma	4	3
FIGO			22.710	0.000*
IIb	24	2
IIIa	18	1
IIIb	7	7
IIIc	6	2
IVa	6	5
IVb	2	5
Size (cm)	4.3 ± 1.3	5.1 ± 2.1	2.212	0.02*
Menopause or not			85.000	0.000*
no	54	10
yes	9	12

Values are given as n or mean ± SD.

*P < 0.05 was considered statistically significant.

BMI, body mass index.

Comparison of R2* and ΔR2* between cervical cancer tumor tissue and normal uterine muscle tissue

The BOLD scanning has a short scanning time(1min and 50 s), stable image quality, and high spatial resolution. Tumor tissue in the color-coded R2* maps, relative to the normal myometrium, reflected a low signal and obvious contrast, enabling clear positioning and measurement (Fig. 1).

The baseline R2* value of cervical cancer tumor and normal uterine muscle tissue followed the normal distribution, and the baseline R2* value of tumor tissue was lower than that of normal uterine muscle tissue (19.67 ± 3.71 Hz and 34.55 ± 3.98 Hz, respectively). The difference was statistically significant (t = 8.162; P < 0.0001). After oxygen stimulation, the signal difference was consistent with the image. The baseline R2* value of the tumor tissue decreased (19.67 ± 3.71 Hz vs. 16.91 ± 3.82 Hz), and the difference was statistically significant (t = 3.161; P = 0.012) (Fig. 2). The baseline R2* value of the tumor tissue area after treatment was higher than that before treatment (Fig. 3), and the difference was statistically significant (t = 4.787; P = 0.007). The regional dynamic △R2* value of the tumor tissue had decreased compared with before treatment, and the difference was statistically significant (t = 2.552; P = 0.025) (Table 2).

Table 2.

Comparison of baseline R2* values and oxygen stimulation dynamic △R2* values of cervical cancer before and after treatment.

	Before treatment	After treatment	t	P
Baseline R2* (Hz)	19.67 ± 3.71	24.63 ± 4.65	4.787	0.007*
Dynamic △R2* (Hz)	2.76 ± 1.65	0.97 ± 0.87	2.552	0.025

Values are given as mean ± SD. Baseline R2* value is R2* values of tumor during air inhalation; △R2* value is the change in R2* value in the same area of tumor before and after oxygen stimulation.

*P < 0.05 was considered statistically significant.

Comparison of R2* and ΔR2* in distinct groups of cervical cancer after treatment

The comparison of baseline R2* and the oxygen stimulation dynamic △R2* of advanced cervical cancer after treatment with different curative effects indicated some differences. The difference in the baseline R2* value was statistically significant (F = 5.144; P = 0.000), and the baseline R2* value of the complete response was the highest after treatment. The difference in dynamic △R2* value was statistically signiﬁcant (F = 3.412; P = 0.017), and the dynamic △R2* value in the complete response was the lowest (Fig. 4 and Table 3).

Table 3.

Comparison of baseline R2* values and oxygen stimulation dynamic △R2* values of cervical cancer tissues after treatment.

	Progression (n=22)	Partial response (n=34)	Complete response (n=29)	F	P
Baseline R2* (Hz)	17.18 ± 2.32	19.66 ± 3.81	21.79 ± 4.12	5.144	0.000*
Dynamic △R2* (Hz)	2.17 ± 1.14	1.34 ± 1.08	0.76 ± 0.21	3.412	0.017

Values are given as mean ± SD. Baseline R2* values were the baseline R2* of tumor during air inhalation; △R2* values were the change in of R2* values before and after oxygen stimulation.

*P < 0.05 was considered statistically significant.

Discussion

Tumor hypoxia indicated an imbalance in oxygen supply and consumption in tumor tissue, a characteristic manifestation of solid tumor microenvironments (4,14,15). Pathology confirmed the existence of tumor hypoxia, indicating that hypoxia may be related to tumor malignancy, apoptosis, tumor angiogenesis, and a tendency to metastasize. The tumor angiogenesis induced and regulated by hypoxia often produces defects in the new blood vessel structure, disabling it to supply oxygen normally and further aggravating tissue hypoxia (16,17). Acute hypoxia is closely related to tumor microcirculation perfusion and has important clinical significance because it can be reversed (14,15,18). The studies showed that the state of tumor oxygen partial pressure changes during treatment and reoxygenation was important for selecting the tumor treatment time (8,19). Therefore, comprehensive and dynamic monitoring of tumor hypoxia can help to effectively evaluate the efficacy and guide the formulation of treatment plan.

BOLD-MRI is affected by the content of deoxyhemoglobin in blood and tissue. Deoxyhemoglobin as a paramagnetic substance can increase the R2* value of blood vessels and surrounding tissue due to the free diffusion of soluble oxygen between perfusion vessels and their adjacent tissue (10,20). Therefore, BOLD-MRI is sensitive to intravascular oxygen partial pressure. BOLD-MRI is affected by the interaction of tissue blood perfusion (blood volume) and erythrocyte oxygenation (12).

The BOLD parameters (T2*/R2* values) are related to the concentration of deoxyhemoglobin in local tissue, which can indirectly reflect the oxygen distribution state of the tissue. A high R2* value indicates a high local deoxyhemoglobin concentration, poor local oxygenation, and tissue in a hypoxic state (8,10 –12,21). High concentration oxygen can improve the dissociation of oxygen in capillary terminals and tissue spaces and cause the ratio of oxygenated-to-deoxyhemoglobin to increase. The right shift of oxygen and hemoglobin dissociation curves can weaken the T2* shortening effect and enhance T2* signal intensity. According to the research (20,22), a change in the R2* values (△R2* values) of tumor caused by hyperoxia mixed gas can indirectly reflect changes in local oxygen partial pressure. A decrease in the R2* values can reflect a decrease in deoxyhemoglobin concentration and an increase in local oxygen partial pressure in tumor tissue (9,11,12). Some previous studies have used the △R2* values of BOLD-MRI to study the internal characteristics of animal and human tumors (20,22). The inhalation gas that is commonly used comprises carbon dioxide and oxygen mixed with a carbon–gold hybrid nanozyme. However, it is often difficult to use carbon–gold gas to study the internal characteristics of human tumors because patients may be unwilling to continue after inhaling carbon–gold gas. This has prompted researchers to use and further develop the clinical signifificance of the BOLD-MRI baseline R2* value parameter (6,9,10,11).

Several studies have investigated the use of BOLD-MRI to noninvasively assess hypoxia of different tumors (23 –26). The imagings of cervical cancer tumor and normal uterine myometrial tissue were compared, and the difference between R2* values was statistically significant. The R2* values of the tumor decreased after inhalation of high concentration oxygen. To some extent, the dynamic △R2* values reflected the response function of tumor tissue to oxygen stimulation, as well as the acute hypoxic state related to microcirculation perfusion (e.g. tumor neovascularization).

The hypoxic status before tumor treatment canbe used as an independent predictor of curative effect (2 –5,7,16 –18). Quantitative hypoxic function parameters, the baseline R2* value, and dynamic oxygen stimulation △R2* value can indirectly reflect tumor hypoxia and response to oxygen stimulation. The results of this study showed there was a significant difference between the baseline R2* and the dynamic △R2* values of oxygen stimulation after treatment among the groups of different therapeutic effects. The results showed that the baseline R2* and the dynamic △R2* values of oxygen stimulation had a degree of curative effect on judgment and potential predictive value. However, since the current study was a single-center study and did not consider the influences of factors such as the number of samples included in our study was small and the follow-up period of patients was only 6 months. Whether the above MR functional parameters could be used as an index for accurately and indirectly reflecting the tumor oxygen partial pressure as a factor for predicting the curative effect thus requires further study.

In conclusion, the BOLD-MRI is easy to operate and has no adverse events. It can quantitatively and non-invasively distinguish cervical cancer from normal cervical tissue and has a degree of curative effect on judgment and prediction values.

Footnotes

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the value of multi-modality magnetic resonance imaging in the treatment decision and prognosis evaluation of cervical cancer, natural science foundation of Gansu (grant number 21JR1RA086).

ORCID iD

Liang Yin

References

Bray

Ferlay

Soerjomataram

, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394–424.

Yang

, et al. Constructed a novel 5 hypoxia genes signature for cervical cancer. Cancer Cell Int 2021;21:345.

Moon

Petersson

Olcina

MM.

The importance of hypoxia in radiotherapy for the immune response, metastatic potential and FLASH-RT. Int J Radiat Biol. 2022;98(3):439–451.

Hompland

Fjeldbo

Lyng

Tumor Hypoxia as a Barrier in Cancer Therapy: Why Levels Matter. Cancers (Basel). 2021 Jan 28;13(3):499.

Kim

Park

, et al. Blood oxygenation leveldependent MR imaging as a predictor of therapeutic response to concurrent chemoradiotherapy in cervical cancer: a preliminary experience. Eur Radiol 2014;24:1514–1520.

Waller

Onderdonk

Flood

, et al. The clinical utility of imaging methods used to measure hypoxia in cervical cancer. Br J Radiol. 2020 Jul;93(1111):20190640.

Lee

Kim

, et al. Value of blood oxygenation level-dependent MRI for predicting clinical outcomes in uterine cervical cancer treated with concurrent chemoradiotherapy. Eur Radiol. 2019 Nov;29(11):6256–6265.

Virani

Kwon

Zhou

, et al. In vivo hypoxia characterization using blood oxygen level dependent magnetic resonance imaging in a preclinical glioblastoma mouse model. Magn Reson Imaging. 2021 Feb;76:52–60.

Hirsch

Toth

Förschler

, et al. Technical considerations on the validity of blood oxygenation level-dependent-based MR assessment of vascular deoxygenation. NMR Biomed 2014;27: 853–862.

10.

O’Connor

JPB

Robinson

Waterton

. Imaging tumor hypoxia with oxygen-enhanced MRI and BOLD MRI. Br J Radiol 2019;92:20180642.

11.

D'Alonzo

Gill

Rowshanfarzad

, et al. In vivo noninvasive preclinical tumor hypoxia imaging methods: a review. Int J Radiat Biol. 2021;97(5):593–631.

12.

Cao-Pham

Tran-Ly-Binh

Heyerick

A, et al. Combined endogenous MR biomarkers to assess changes in tumor oxygenation induced by an allosteric effector of hemoglobin. NMR Biomed. 2020 Feb;33(2):e4181.

13.

Eisenhauer

Therasse

Bogaerts

, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009 Jan;45(2):228–47.

14.

Liu

Ciotti

Eisinger-Mathason

TSK

. Hypoxia and the tumor secretome. Adv Exp Med Biol 2019;1136:57–69.

15.

Zhao

XF.

Hypoxia and the tumor microenvironment. Technol Cancer Res Treat 2021;20:15330338211036304.

16.

Vaupel

Mayer

Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev 2007;26:225–239.

17.

Abou Khouzam

Brodaczewska

Filipiak

, et al. Tumor Hypoxia Regulates Immune Escape/Invasion: Influence on Angiogenesis and Potential Impact of Hypoxic Biomarkers on Cancer Therapies. Front Immunol. 2021 Jan 20;11:613114.

18.

Hughes

Wiggins

Siemann

DW.

Tumor oxygenation and cancer therapy-then and now. Br J Radiol. 2019 Jan;92(1093):20170955.

19.

Lyng

Sundfør

Rofstad

EK.

Changes in tumor oxygen tension during radiotherapy of uterine cervical cancer: relationships to changes in vascular density, cell density, and frequency of mitosis and apoptosis. Int J Radiat Oncol Biol Phys 2000;46:935–946.

20.

Yang

Arai

Campbell

3rd, et al. Oxygen-sensitive MRI assessment of tumor response to hypoxic gas breathing challenge. NMR Biomed. 2019 Jul;32(7):e4101.

21.

Gallez

The Role of Imaging Biomarkers to Guide Pharmacological Interventions Targeting Tumor Hypoxia. Front Pharmacol. 2022 Jul 15;13:853568.

22.

Landuyt

Hermans

Bosmans

, et al. BOLD contrast fMRI of whole rodent tumor during air or carbogen breathing using echo-planar imaging at 1.5 T. Eur Radiol 2001;11:2332– 2340.

23.

Christen

Lemasson

Pannetier

, et al. Is T2* enough to assess oxygenation? Quantitative blood oxygenation level-dependent analysis in brain tumor. Radiology 2012;262:495–502.

24.

Neugarten

. Renal BOLD-MRI and assessment for renal hypoxia. Kidney Int 2012;81:613–614.

25.

Hoskin

Carnell

Taylor

, et al. Hypoxia in prostate cancer: correlation of BOLD-MRI with pimonidazole immunohistochemistryinitial observations. Int J Radiat Oncol Biol Phys 2007;68:1065–1071.

26.

Huang

, et al.Oxygen-Challenge Blood Oxygen Level-Dependent Magnetic Resonance Imaging for Evaluation of Early Change of Hepatocellular Carcinoma to Chemoembolization: A Feasibility Study. Acad Radiol. 2021 Nov;28 Suppl 1:S13–S19.