Magnetic Resonance Imaging xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Magnetic Resonance Imaging journal homepage: www.mrijournal.com
Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer Zhan-Zhao Fu a,⁎, Yong Peng b, Li-Yan Cao a, Yan-Sheng Chen c, Kun Li d, Bao-Hong Fu a a
Department of Oncology, the First Hospital of Qinhuangdao, Qinhuangdao 066000, P.R. China Department of Bioengineering, Yanshan University, Qinhuangdao 066000, P.R. China Division of NMR Research, the First Hospital of Qinhuangdao, Qinhuangdao 066000, P.R. China d Department of Chemical Engineering, Yanshan University, Qinhuangdao 066000, P.R. China b c
a r t i c l e
i n f o
Article history: Received 17 May 2014 Revised 31 January 2015 Accepted 1 February 2015 Available online xxxx Keywords: Diffusion-weighted MR imaging Apparent diffusion coefficient Cervical cancer Radiotherapy Chemotherapy Meta-analysis
a b s t r a c t Objective: We investigated the clinical significance of apparent diffusion coefficient (ADC) values in diffusion-weighted magnetic resonance imaging (DWI) in monitoring the efficacy of radiotherapy (RT) and chemotherapy (CT) treatments in cervical cancer. Method: In order to identify relevant high quality clinical cohort studies reporting the use of DWI in cervical cancers, the following electronic databases in English and Chinese languages were comprehensively searched: MEDLINE, Science Citation Index database, Cochrane Library Database, PubMed, Embase, CINAHL, and Current Contents Index; Chinese Biomedical Database, Chinese Journal Full-Text Database. All selected studies were published prior to March 2014, and data extracted from these studies were analyzed using STATA 12.0 statistical software. Results: We initially retrieved 196 articles (79 Chinese articles and 117 English articles) through database searches and finally selected sixteen cohort studies for this meta-analysis. The 16 studies contained a combined total of 517 subjects, and all selected studies reported the mean ADC value (10−3 mm2/s) in DWI in cervical cancer patients treated with RT and CT. Combined standardized mean difference (SMD) suggested that the mean post-RT and mean post-CT ADC values were significantly higher than the mean pre-RT and mean pre-CT ADC values, respectively, in cervical cancer patients (SMD = 2.95, 95% CI = 2.19–3.72, P b 0.001). Ethnicity-stratified analysis revealed that increased ADC values were observed post-RT and post-CT in both Caucasian (SMD = 1.44, 95% CI = 0.93–1.95, P b 0.001) and Asian populations (SMD = 3.32, 95% CI = 2.42–4.22, P b 0.001), compared with the mean ADC values before RT and CT, respectively, in the two subgroups. Further, subgroup analysis based on b-value revealed that higher ADC values were found in cervical cancer patients after RT and CT, compared to before RT and CT treatment, with both b value ≤ 900 (SMD = 3.71, 95% CI = 2.35–5.07, P b 0.001) and N900 (SMD = 2.55, 95% CI = 1.78–3.32, P b 0.001). The mean ADC value in patients without residual tumor post-RT and post-CT treatment was significantly higher than seen in patients with residual tumors (SMD = 0.80, 95% CI = 0.49–1.12, P b 0.001). Conclusion: Our meta-analysis revealed a significant correlation between mean ADC values and the clinical response to RT and CT treatment. Thus, ADC values in DWI may be effective in evaluating the clinical outcome of treatments in cervical cancer patients. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Cervical cancer is the second most common gynecological disease and the third most frequent cause of cancer mortality worldwide [1,2]. Cervical cancer predominantly occurs in women between the third and fifth decade of their life [3]. Approximately, 3 out of 10 cervical cancer patients will have tumor recurrence and succumb to ⁎ Corresponding author at: Department of Oncology, the First Hospital of Qinhuangdao, Wenhua Road No.2, Haigang District, Qinhuangdao 066000, P.R. China. Tel./fax: +86 335 3508121. E-mail address: [email protected] (Z-Z. Fu).
the disease, and the 5-year survival rate of cervical cancers is only 50% [4]. Traditional treatment consists of radical hysterectomy with lymph node dissection, and patients with early stage cervical cancers have a better response with this approach [5]. However, in patients with tumors larger than 4 cm in diameter, the prognosis is poor, and survival rates of 50–60% are seen in patients with large tumors [6]. Such poor results are attributed to increased incidence of risk factors, such as vascular space invasion, lymph node metastases, deep cervical invasion, undiagnosed parametrial extension and radioresistant hypoxic tumor centers [7]. In this context, a recent approach of neoadjuvant therapy, i.e., giving chemotherapy or radiotherapy prior to radical surgery, might shrink tumors, making it easier for surgical
http://dx.doi.org/10.1016/j.mri.2015.02.002 0730-725X/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
2
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
removal and also help destroy tiny tumors or tumor spreads that go undetected [8]. Currently, radiotherapy (RT) and chemotherapy (CT) remain standard therapeutic approaches for cervical cancers and significantly improve survival, and reduce local and distant recurrence [9]. The newer approaches for successful treatment of cancers rely heavily on diagnostic methods to detect cancers early and involve monitoring treatment efficacy. In cervical cancer setting, there is an urgent need for reliable indicators to measure therapeutic response to minimize adverse side effects and to improve quality of life. For this reason, diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) measurement is an attractive diagnostic tool for cervical cancer patients [10,11]. Diffusion-weighted magnetic resonance imaging (DWI) is highly sensitive to cellularity, viscosity and extracellular free water [12,13]. By comparison with other imaging methods, DWI is completely noninvasive and does not involve exposure to ionizing radiation [14]. In addition, DWI does not require contrast medium and permits more frequent examinations, allowing for repeated monitoring throughout the cancer treatment regimen [3]. DWI exploits the physical principle that the degree of water diffusion in biological tissues is determined by tissue cellularity and the integrity of cell membranes [15]. The random Brownian motion of water protons through biological tissues is impeded by the tissue architecture, and normal tissues have low cell numbers and larger extracellular spaces, which allows greater mobility of molecular water. On the other hand, tumors at various stages exhibit increasingly complex architecture,
with densely packed cells, altered membrane permeability and highly restricted extracellular space, all of which impede water mobility. Importantly, ADC value, a quantitative parameter in DWI that measures the rate of water mobility, increases after successful therapy because tumor cell density decreases, resulting in removal of barriers to water motion [16,17]. In clinic, ADC values are vital for planning therapy, in identification of tumor boundaries and in deciding eligibility for surgical resection [18]. Recently, DWI has also been used successfully to evaluate the therapeutic efficacy of non-surgical treatments in breast cancer, bladder cancer and rectal cancer, and avoid radical surgery to prevent unnecessary removal of the affected organ [19–21]. In the context of this study, ADC values in DWI may be helpful in evaluating the efficacy of concurrent chemo-radiotherapy treatment in cervical cancers [11]. Accumulated evidence indicates that DWI can be used as a surrogate biomarker to assess tumor cellularity in cervical cancers [3,22], but contradictory results are also reported [4,23]. Therefore, we analyzed published data by meta-analysis to assess the clinical value of DWI with ADC values in evaluation of response to RT and CT in cervical cancer patients. 2. Materials and methods 2.1. Literature search and data sources Relevant articles were searched using the following bibliographic databases: MEDLINE (Ovid version, New York, 1966 ~ 2013), Science
Fig. 1. Flow chart of literature search and study selection.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
3
Table 1 Main characteristics and methodological quality of eligible studies. First author
Year
Country
Language
Ethnicity
Patients
Age (years)
MRI machine type
b value (s/mm2)
Kuang F. [37] Lu X. H. [32] Gao Y. [29] Cui G. [28] Nakamura K. [2] Makino H. [38] Kim H. S. [22] Hu A. M. [30] Somoye G. [4] Shi Y. B. [36] Zhang Y. [10] Rizzo S. [35] Cao K. [39] Ma W. L. [33] Liu Y. [3] Harry V. N. [17]
2014 2013 2013 2013 2013 2013 2013 2012 2012 2011 2011 2011 2009 2009 2009 2008
China China China China Japan Japan Korea China UK China China Italy China China China UK
English Chinese Chinese Chinese English English English Chinese English Chinese English English Chinese Chinese English English
Asians Asians Asians Asians Asians Asians Asians Asians Caucasians Asians Asians Caucasians Asians Asians Asians Caucasians
75 31 45 45 69 25 24 50 20 18 14 17 21 16 17 20
49.39 (36 ~ 66) 59.4 (39 ~ 73) 43 (26 ~ 71) 49.6 (39 ~ 68) 62 (31 ~ 90) 63.5 (36 ~ 91) 57.3 (35 ~ 76) 56.3 ± 11.8 (30 ~ 76) 45.3 (29.8 ~ 60.8) 46.1 ± 11.3 (31 ~ 71) 44 (26 ~ 71) 50 (35 ~ 75) 48.6 ± 9.3 (33 ~ 70) 52 (43 ~ 62) 49 25 ~ 69) 50 (34 ~ 80)
Siemens 3.0 T GE 1.5 T GE 3.0 T GE 1.5 T Siemens 1.5 T Philips 1.5 T Philips 3.0 T Philips 1.5 T GE 1.5 T Philips 1.5 T GE 1.5 T Siemens 1.5 T GE 1.5 T Siemens 3.0 T GE 1.5 T GE 1.5 T
600 1000 700 700 1000 1000 1000 600 1000 1000 1000 900 1000 1000 1000 1000
ADC = apparent diffusion coefficient.
Citation Index database (Web of Science, Thomson ISI, Philadelphia, PA, USA, 1945 ~ 2013), the Cochrane Library Database (Oxford, UK, Issue 12, 2013), PubMed (National Library of Medicine, Bethesda, MD, USA, 1966 ~ 2013), Embase (Data Star version, Cary, NC, 1974 ~ 2013), CINAHL (CD-ROM Information Retrieval System, 1982 ~ 2013), and Current Contents Index (Institute for Scientific Information, 1995 ~ 2013). We also searched three Chinese databases (Chinese Biomedical
Database, 1978 ~ 2013; the Chinese Journal Full-Text Database, 1980 ~ 2013; and the Weipu Journal Database, 1989 ~ 2013) to identify articles published in Chinese language. We used a combination of the following medical subject and free language terms in a highly sensitive search strategy: (“Diffusion Magnetic Resonance Imaging” or “Diffusion MRI” or “Diffusion Weighted MRI” or “DWI” or “MRI-DWI” or “diffusion-weighted imaging” or “diffusion-weighted-MRI”) and
Fig. 2. NOS scoring of included studies.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
4
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
Fig. 3. Forest plots for the differences in diffusion-weighted magnetic resonance imaging for predicting the efficacy of radiotherapy and chemotherapy for cervical cancer (A: posttreatment vs. pretreatment, B: no residue tumor vs. residue tumor).
(“Chemoradiotherapy” or “radiochemotherapy” or “concurrent chemoradiotherapy” or “synchronous chemoradiotherapy” or “concomitant radiochemotherapy” or “CTRT” or “CT-RT” or “CC-RT”) and (“Uterine Cervical Neoplasms” or “Cervical Neoplasms” or “Cervical cancer” or “cervical carcinoma” or “cervix cancer” or “Cervix Neoplasms” or “Cervix carcinoma”). We combined results of electronic searches with further manual searches to identify other potential articles from cross-references.
2.2. Selection criteria Studies reported in languages unfamiliar to the authors were translated. Full-text of the papers was retrieved if at least one study investigator considered the abstract suitable. Investigators independently assessed the retrieved studies based on the following inclusion criteria: (1) studies must be a clinical cohort study reporting the efficacy of mean ADC values to predict the tumor response to RT and CT treatment of cervical cancer; (2) all patients with pathologically confirmed cervical cancer underwent diffusionweighted magnetic resonance imaging (DW-MRI) before and during RT and CT treatment; (3) the study must contain sufficient information on the mean ADC value for assessing response before and after RT and CT treatment for cervical cancer. Certain publication types, such as abstracts, letters to the editor, and proceedings from scientific meetings and the articles that did not satisfy the current inclusion criteria were excluded. The most recent or the largest
sample size publication was selected when studies using the same subjects were published.
2.3. Data extraction and quality assessment Two investigators independently assessed the data in each study. Disagreements in study selection or data collection were resolved by consensus, and a third reviewer was consulted if necessary. We used a piloted data-extraction sheet, which covered the following descriptive information from included studies: surname and initials of the first author, the year of publication or submission, source country and ethnicity, language of publication, study design, sample size, demographic variables of the subjects, therapy type, MRI type, MRI machine type code, b-value (s/mm 2) and the mean ADC value (10 −3 mm 2/s) before and after RT and CT treatment. The quality of the trials was assessed independently by the investigators using Newcastle-Ottawa Scale (NOS) (http://www.ohri.ca/programs/ clinical_epidemiology/oxford.asp). According to NOS, 9 entries are listed as follows: (1) selection: representativeness of the exposed cohort (NOS01), if exposed cohort and non-exposed cohort come from the same population (NOS02), ascertainment of exposure (NOS03), demonstration that outcome of interest was not present at start of study (NOS04); (2) comparability: control of important factors (NOS05), control of other important mixed factors (NOS06); (3) outcome: independent studies and blinded method (NOS07), was follow-up long enough for outcomes to occur (NOS08), assesses
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
5
A
B
Fig. 4. Subgroup analysis by ethnicity and b value for the differences in diffusion-weighted magnetic resonance imaging for predicting the efficacy of radiotherapy and chemotherapy for cervical cancer (A: ethnicity, B: b value).
the follow-up of the cohorts to ensure that losses are not related to either the exposure or the outcome (NOS9).
2.4. Statistical analysis Standardized mean difference (SMD) was chosen as the summary statistic for meta-analysis and computed with its corresponding 95% confidence interval (95% CI). Pooled estimates of the statistical significance of SMDs were obtained from the Z test. The heterogeneity between trials for each comparison was estimated by use of Cochran's Q-statistic and I 2 tests [24]. If Q-test showed evidence of a P b 0.05 or I 2 test exhibited N 50%, meta-analysis was done using a random-effects model, otherwise fixed-effects model was used [25,26]. Furthermore, univariate and multivariate meta-regression analysis was used to identify heterogeneity, and Monte Carlo simulation was used for multiple testing corrections. A sensitivity analysis was employed to evaluate the impact of single studies on
the overall estimate. To test potential for publication bias across included trials, we created funnel plots and undertook the Egger's linear regression test [27]. All P values were two-tailed, and values P b 0.05 were considered as statistically significant. Statistical analyses were conducted with the STATA statistical software (Version 12.0, Stata Corporation, and College Station, TX, USA). 3. Results 3.1. Description of included studies Fig. 1 shows the flow diagram of study selection process for this meta-analysis. Initially, we retrieved 196 articles by electronic database search. After screening titles and abstracts of all the retrieved articles, we systematically reviewed 101 full text articles, and identified 16 articles based on a careful assessment of quality, relevance and completeness of data [2–4,10,22,23,28–39]. The included studies consisted of a combined total of 517 patients
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
6
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
C
D
E
F
Fig. 5. Meta-regression analysis for the differences in diffusion-weighted magnetic resonance imaging for predicting and assessing cervical cancer (A: publication year, B: sample size, C: b value, D: instrument, E: ethnicity, F: language).
with cervical cancers, with sample size ranging from 28 to 150 between the studies. Of the 16 studies included in the analysis, 13 were conducted in Asians (ten from China, two from Japan, and one from Korea), and the other three were performed in Caucasians (two from UK, and one from Italy). The MRI machine types included with GE 1.5/3.0 T MRI scanners, 1.5 T/3.0 Siemens Avanto scanners, Philips 1.5/3.0 T scanners and Tesla 3.0 T scanners. The mean ADC value before and after RT and CT treatments were obtained. Characteristics of included studies are presented in Table 1, and the NOS assessments for quality are displayed in Fig. 2, which indicated a moderate to high study quality.
Table 2 Meta-regression analyses of potential source of heterogeneity. Heterogeneity factors
Coefficient
Year Sample b value Machine Ethnicity Language
0.194 −0.01 −0.003 0.12 −0.686 1.611
SE
0.339 0.02 0.005 0.275 1.453 1.132
t
0.57 −0.52 −0.68 0.44 −0.47 1.42
P (adjusted)
0.982 0.987 0.959 0.997 0.994 0.555
SE = standard error; LL = lower limit; UL = upper limit.
95% CI LL
UL
−0.572 −0.054 −0.014 −0.501 −3.973 −0.949
0.96 0.034 0.007 0.741 2.601 4.171
3.2. Meta-analysis A total of 16 studies reported differences in the mean ADC values in cervical cancer patients before and after RT and CT treatment. The result of heterogeneity test suggested presence of heterogeneity among the included studies (I 2 = 94.5%, Ph b 0.001), therefore a random-effects model was chosen. The result of meta-analysis showed that the mean post-RT and post-CT ADC values were significantly higher than the mean pre-RT and pre-CT ADC values in cervical cancer patients (SMD = 2.95, 95% CI = 2.19–3.72, P b 0.001) (Fig. 3A). Six studies reported the differences in the mean ADC values in patients with residual tumor and in patients with no residual tumor post-RT and post-CT treatment. Heterogeneity test results suggest that there was no heterogeneity among the included studies (I 2 = 0.0%, Ph = 0.640), thus a fixed-effects model was chosen. Meta-analysis of the results showed that the mean ADC values in patients with no residual tumors was significantly higher than the ADC values observed in patients with residual tumors post-RT and post-CT treatment (SMD = 0.80, 95% CI = 0.49–1.12, P b 0.001) (Fig. 3B). In order to explore potential sources of heterogeneity, we also performed subgroup analyses based on ethnicity and b value. Ethnicitystratified analysis demonstrated that both Caucasians (SMD = 1.44, 95% CI = 0.93–1.95, P b 0.001) and Asians (SMD = 3.32, 95% CI = 2.42–
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
C
D
7
Fig. 6. Sensitivity analysis and funnel plot of publication biases for the differences in diffusion-weighted magnetic resonance imaging for predicting and assessing cervical cancer (A–B:. sensitivity analysis, C–D: publication bias).
4.22, P b 0.001) (Fig. 4A) showed a similar increase in ADC values post-RT and post-CT, compared with ADC values pre-RT and pre-CT. Further, subgroup analysis based on b-value revealed significantly higher ADC values in cervical cancer patients post-RT and post-CT treatment compared to pre-RT and pre-CT treatment with both b value ≤ 900 (SMD = 3.71, 95% CI = 2.35–5.07, P b 0.001) and N 900 (SMD = 2.55, 95% CI = 1.78–3.32, P b 0.001) (Fig. 4B). Univariate regression analysis showed language as a major source of heterogeneity (P = 0.024), while publication year, sample size, b value, instruments and ethnicity showed no heterogeneity (publication year: P = 0.338; sample size: P = 0.461; b value: P = 0.143; instrument: P = 0.422; ethnicity: P = 0.113) (Fig. 5). Multivariate regression analysis showed that publication year, sample size, b value, instrument, ethnicity, language were not heterogeneity sources (Table 2). 3.3. Sensitive analysis and publication bias Sensitivity analysis showed that all included studies had no effect on the pooled SMD of the mean ADC values in patients with residual tumors and no residual tumors pre- and post-RT and CT treatment (Fig. 6A–B). Funnel plot analysis in relation to the differences of the mean ADC values revealed evidence of obvious asymmetry in cervical cancer patients before and after RT and CT treatment (Fig. 6C). Further Egger's linear regression analysis confirmed the presence of publication bias of the included studies (P = 0.032). However, the funnel plot analysis revealed no evidence of asymmetry in cervical cancer patients with residual tumors and no residual tumors (Fig. 6D), which was further confirmed by Egger's linear regression analysis (P = 0.609).
4. Discussion The present meta-analysis was carried out to assess the efficacy of DWI in evaluating clinical response to RT and CT in cervical cancer patients. ADC measurements are useful for early detection, determination of malignancy and evaluation of treatment response [40]. In general, high ADC values convey that water molecules can move freely in the tissue, suggesting low cellularity and better organization of tissue structure, a typical feature of normal tissues. On the other hand, low ADC values indicate impeded mobility of water molecules, suggesting high cellularity, a typical feature in malignant lesions. Benign lesions, including simple cysts and hemangiomas, also have higher ADC values due to low cell density [41–43]. The major result of our meta-analysis revealed that ADC values of cervical cancer patients post-RT and CT were significant higher than pre-RT and CT, and the ADC values of patients with no residual tumors was significantly higher than patients with residual tumors after RT and CT, indicating that DWI may be effective in evaluating the clinical response to RT and CT. DWI is highly sensitivity to the motion of water molecules and permits non-invasive characterization of biological tissues based water diffusion properties [44,45]. Through the application of diffusion-weighted gradient pulses to conventional magnetic resonance sequence, the sensitivity of the signal to the level of localized water diffusibility, quantified as ADC, can be achieved. ADC obtained from DWI is a novel assessment tool and functions as a biomarker for assessing early response to treatment in various cancer types [23]. ADC can provide useful information on tumor aggressiveness, tumor cellularity, subtype characterization and clinical response to cancer treatment [46]. A significant change
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
8
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
in ADC values after treatment with RT and CT is regarded as a positive clinical response, while absence of change in ADC values indicates disease progression [47]. Since our results show a significant difference in ADC values pre- and post-RT and CT, we postulate that DWI is effective in evaluating clinical response to RT and CT in cervical cancer patients. Rizzo et al. also observed that ADC values were significantly different between normal and cancer tissue in the cervix, and between cervical tumors before and after nonsurgical therapy in all patients studied [35]. In a study by Makino et al., DWI and ADC measurement were useful in early evaluation of response to RT and CT in cervical cancer patients [34]. Our stratified analysis based on ethnicity and b-value excluded the potential of heterogeneity based on race. In the ethnicitystratified analysis, we found that the ADC values were significantly increased after RT and CT treatment in both Asian and Caucasian populations. Furthermore, the results of subgroup analysis by b-value suggested that there was significant difference in ADC values before and after RT and CT treatment in majority of subgroups, indicating that selection of b-value may not have an impact on the overall results. In summary, our meta-analysis results are consistent with previous studies that showed an elevation in ADC values in cervical cancer patients after RT and CT treatment, supporting the view that DWI is a reliable surrogate biomarker for assessing clinical responses in cervical cancer patients. Our meta-analysis has potential limitations. First, potential sources of bias must be considered: all studies included in our meta-analysis had methodological limitations such as no intentionto-treat analysis, unclear or inadequate data collection. Study limitations also include a small sample size and lack of a uniform standard on ADC values in judging the efficacy of cervical cancer treatment. African populations were not included in this analysis, which may result in publication bias. Second, the studies could be significantly heterogeneous because some of the planned analyses were restricted by small sample size within those trials. A third limitation in our meta-analysis is that we did not consider articles and abstracts. Therefore, the results of this meta-analysis must be interpreted with caution. Despite the above limitations, this is the first example of a meta-analysis to test the association of ADC values in DWI with the clinical response in cervical cancer. In conclusion, our meta-analysis provides evidence that increase in ADC values of DWI is a reliable indicator of favorable treatment response in cervical cancer patients. Nevertheless, due to the limitations of our study, prospective studies are needed in the future to obtain a standardized assessment of diagnostic markers and confirm our findings in a larger patient population. Conflicts of interest No competing financial interests exist. Acknowledgments We would like to acknowledge the reviewers for their helpful comments on this paper. References [1] Bygbjerg IC. Double burden of noncommunicable and infectious diseases in developing countries. Science 2012;337:1499–501. [2] Nakamura K, Kajitani S, Joja I, Haruma T, Fukushima C, Kusumoto T, et al. The posttreatment mean apparent diffusion coefficient of primary tumor is superior to pretreatment ADCmean of primary tumor as a predictor of prognosis with cervical cancer. Cancer Med 2013;2:519–25. [3] Liu Y, Bai R, Sun H, Liu H, Zhao X, Li Y. Diffusion-weighted imaging in predicting and monitoring the response of uterine cervical cancer to combined chemoradiation. Clin Radiol 2009;64:1067–74.
[4] Somoye G, Harry V, Semple S, Plataniotis G, Scott N, Gilbert FJ, et al. Early diffusion weighted magnetic resonance imaging can predict survival in women with locally advanced cancer of the cervix treated with combined chemoradiation. Eur Radiol 2012;22:2319–27. [5] Pfaendler KS, Wenzel L, Mechanic MB, Penner KR. Cervical cancer survivorship: long-term quality of life and social support. Clin Ther 2015;37:39–48. [6] Lintner B, Saso S, Tarnai L, Novak Z, Palfalvi L, Del Priore G, et al. Use of abdominal radical trachelectomy to treat cervical cancer greater than 2 cm in diameter. Int J Gynecol Cancer 2013;23:1065–70. [7] Wang N, Guan QL, Wang K, Zhou X, Gao C, Yang HT, et al. Radiochemotherapy versus radiotherapy in locally advanced cervical cancer: a meta-analysis. Arch Gynecol Obstet 2011;283:103–8. [8] Rydzewska L, Tierney J, Vale CL, Symonds PR. Neoadjuvant chemotherapy plus surgery versus surgery for cervical cancer. Cochrane Database Syst Rev 2012;12:CD007406. [9] Katanyoo K, Sanguanrungsirikul S, Manusirivithaya S. Comparison of treatment outcomes between squamous cell carcinoma and adenocarcinoma in locally advanced cervical cancer. Gynecol Oncol 2012;125:292–6. [10] Zhang Y, Chen JY, Xie CM, Mo YX, Liu XW, Liu Y, et al. Diffusion-weighted magnetic resonance imaging for prediction of response of advanced cervical cancer to chemoradiation. J Comput Assist Tomogr 2011;35:102–7. [11] Levy A, Caramella C, Chargari C, Medjhoul A, Rey A, Zareski E, et al. Accuracy of diffusion-weighted echo-planar MR imaging and ADC mapping in the evaluation of residual cervical carcinoma after radiation therapy. Gynecol Oncol 2011;123: 110–5. [12] Zhang J, Tehrani YM, Wang L, Ishill NM, Schwartz LH, Hricak H. Renal masses: characterization with diffusion-weighted MR imaging–a preliminary experience. Radiology 2008;247:458–64. [13] Kim CK, Park BK, Kim B. Diffusion-weighted MRI at 3 T for the evaluation of prostate cancer. AJR Am J Roentgenol 2010;194:1461–9. [14] Sugita R, Yamazaki T, Furuta A, Itoh K, Fujita N, Takahashi S. High b-value diffusion-weighted MRI for detecting gallbladder carcinoma: preliminary study and results. Eur Radiol 2009;19:1794–8. [15] Razek AA, Megahed AS, Denewer A, Motamed A, Tawfik A, Nada N. Role of diffusion-weighted magnetic resonance imaging in differentiation between the viable and necrotic parts of head and neck tumors. Acta Radiol 2008;49:364–70. [16] Song I, Kim CK, Park BK, Park W. Assessment of response to radiotherapy for prostate cancer: value of diffusion-weighted MRI at 3 T. AJR Am J Roentgenol 2010;194:W477–82. [17] Harry VN. Novel imaging techniques as response biomarkers in cervical cancer. Gynecol Oncol 2010;116:253–61. [18] McVeigh PZ, Syed AM, Milosevic M, Fyles A, Haider MA. Diffusion-weighted MRI in cervical cancer. Eur Radiol 2008;18:1058–64. [19] Park SH, Moon WK, Cho N, Song IC, Chang JM, Park IA, et al. Diffusion-weighted MR imaging: pretreatment prediction of response to neoadjuvant chemotherapy in patients with breast cancer. Radiology 2010;257:56–63. [20] Yoshida S, Koga F, Kobayashi S, Ishii C, Tanaka H, Tanaka H, et al. Role of diffusion-weighted magnetic resonance imaging in predicting sensitivity to chemoradiotherapy in muscle-invasive bladder cancer. Int J Radiat Oncol Biol Phys 2012;83:e21–7. [21] Jung SH, Heo SH, Kim JW, Jeong YY, Shin SS, Soung MG, et al. Predicting response to neoadjuvant chemoradiation therapy in locally advanced rectal cancer: diffusion-weighted 3 Tesla MR imaging. J Magn Reson Imaging 2012;35:110–6. [22] Kim HS, Kim CK, Park BK, Huh SJ, Kim B. Evaluation of therapeutic response to concurrent chemoradiotherapy in patients with cervical cancer using diffusionweighted MR imaging. J Magn Reson Imaging 2013;37:187–93. [23] Harry VN, Semple SI, Gilbert FJ, Parkin DE. Diffusion-weighted magnetic resonance imaging in the early detection of response to chemoradiation in cervical cancer. Gynecol Oncol 2008;111:213–20. [24] Zintzaras E, Ioannidis JP. HEGESMA: genome search meta-analysis and heterogeneity testing. Bioinformatics 2005;21:3672–3. [25] Zintzaras E, Ioannidis JP. Heterogeneity testing in meta-analysis of genome searches. Genet Epidemiol 2005;28:123–37. [26] Zhou L, Yao F, Luan H, Wang Y, Dong X, Zhou W, et al. Three novel functional polymorphisms in the promoter of FGFR2 gene and breast cancer risk: a HuGE review and meta-analysis. Breast Cancer Res Treat 2012;136:885–97. [27] Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Comparison of two methods to detect publication bias in meta-analysis. JAMA 2006;295:676–80. [28] Cui G, Yang LJ, Zhang YP, Guo XF, Wang DK, Shi Y. Diffusion weighted imaging in predicting and assessing therapeutic effect of cervical carcinoma. Chin J Clin (Electron Ed) 2013;7:7573–5. [29] Gao Y, Wang F, Kong LL, Wu LM. Diffusion-weighted imaging in predicting and monitoring the response of advanced cervical cancer to combined radiotherapy. Anhui Med Pharm J 2013;17:967–9. [30] Hu AM, Wang Y, X., Zhou SJ, Zhu ZW, Huang ZH, et al. The applied research of diffusion weighted imaging in cervical carcinoma. J Wenzhou Med Coll 2012;42:474–7. [31] Li YL. The applied research of DWI in monitoring therapeutic effect of uterine cervix cancer with radiotherapy. Xinli Yesheng; 2012 114. [32] Lu XH, Wang J, Liu WY, Ding Y. Diffusion weighted imaging in predicting and assessing therapeutic effect of cervical carcinoma. J Chin Pract Diagn Ther 2013;27:358–60. [33] Ma WL, Huan Y, Wei LC, Lu J, Xu JQ, Zhao HT, et al. The applied research of 3.0 T MRI and DWI in monitoring therapeutic effect of uterine cervix cancer with radiotherapy. J Clin Radiol 2009;28:1651–4. [34] Makino H, Kato H, Furui T, Morishige KI, Kanematsu M. Predictive value of diffusion-weighted magnetic resonance imaging during chemoradiotherapy for uterine cervical cancer. J Obstet Gynaecol Res 2014;40:1098–104.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx [35] Rizzo S, Summers P, Raimondi S, Belmonte M, Maniglio M, Landoni F, et al. Diffusion-weighted MR imaging in assessing cervical tumour response to nonsurgical therapy. Radiol Med 2011;116:766–80. [36] Shi YB, Tong HZ, Chen MD, Zhou Q. Evaluation of the early chemoradiation efficacy in cervical cancer by diffusion weighted magnetic resonance imaging. J Med Imaging 2011;21:887–9. [37] Kuang F, Yan Z, Wang J, Rao Z. The value of diffusion-weighted MRI to evaluate the response to radiochemotherapy for cervical cancer. Magn Reson Imaging 2014;32: 342–9. [38] Makino H, Kato H, Furui T, Morishige KI, Kanematsu M. Predictive value of diffusion-weighted magnetic resonance imaging during chemoradiotherapy for uterine cervical cancer. J Obstet Gynaecol Res 2013;40:1098–104. [39] Cao K, Zhang XP, Wang N, Shan J, Cui Y, Qi LP. Evaluation of the earl y chemoradiation efficacy in cervical cancer by diffusion-weighted magnetic resonance imaging. Chin J Med Imaging Technol 2009;1657–60. [40] Klimas A, Drzazga Z, Kluczewska E, Hartel M. Regional ADC measurements during normal brain aging in the clinical range of b values: a DWI study. Clin Imaging 2013;37:637–44.
9
[41] Testa ML, Chojniak R, Sene LS, Damascena AS, Guimaraes MD, Szklaruk J, et al. Is DWI/ADC a useful tool in the characterization of focal hepatic lesions suspected of malignancy? PLoS One 2014;9:e101944. [42] Li SP, Padhani AR. Tumor response assessments with diffusion and perfusion MRI. J Magn Reson Imaging 2012;35:745–63. [43] Galea N, Cantisani V, Taouli B. Liver lesion detection and characterization: role of diffusion-weighted imaging. J Magn Reson Imaging 2013;37:1260–76. [44] Padhani AR, Liu G, Koh DM, Chenevert TL, Thoeny HC, Takahara T, et al. Diffusionweighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia 2009;11:102–25. [45] Hamstra DA, Rehemtulla A, Ross BD. Diffusion magnetic resonance imaging: a biomarker for treatment response in oncology. J Clin Oncol 2007; 25:4104–9. [46] Koh DM, Collins DJ. Diffusion-weighted MRI in the body: applications and challenges in oncology. AJR Am J Roentgenol 2007;188:1622–35. [47] Moffat BA, Chenevert TL, Lawrence TS, Meyer CR, Johnson TD, Dong Q, et al. Functional diffusion map: a noninvasive MRI biomarker for early stratification of clinical brain tumor response. Proc Natl Acad Sci U S A 2005;102:5524–9.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Contents lists available at ScienceDirect
Magnetic Resonance Imaging journal homepage: www.mrijournal.com
Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer Zhan-Zhao Fu a,⁎, Yong Peng b, Li-Yan Cao a, Yan-Sheng Chen c, Kun Li d, Bao-Hong Fu a a
Department of Oncology, the First Hospital of Qinhuangdao, Qinhuangdao 066000, P.R. China Department of Bioengineering, Yanshan University, Qinhuangdao 066000, P.R. China Division of NMR Research, the First Hospital of Qinhuangdao, Qinhuangdao 066000, P.R. China d Department of Chemical Engineering, Yanshan University, Qinhuangdao 066000, P.R. China b c
a r t i c l e
i n f o
Article history: Received 17 May 2014 Revised 31 January 2015 Accepted 1 February 2015 Available online xxxx Keywords: Diffusion-weighted MR imaging Apparent diffusion coefficient Cervical cancer Radiotherapy Chemotherapy Meta-analysis
a b s t r a c t Objective: We investigated the clinical significance of apparent diffusion coefficient (ADC) values in diffusion-weighted magnetic resonance imaging (DWI) in monitoring the efficacy of radiotherapy (RT) and chemotherapy (CT) treatments in cervical cancer. Method: In order to identify relevant high quality clinical cohort studies reporting the use of DWI in cervical cancers, the following electronic databases in English and Chinese languages were comprehensively searched: MEDLINE, Science Citation Index database, Cochrane Library Database, PubMed, Embase, CINAHL, and Current Contents Index; Chinese Biomedical Database, Chinese Journal Full-Text Database. All selected studies were published prior to March 2014, and data extracted from these studies were analyzed using STATA 12.0 statistical software. Results: We initially retrieved 196 articles (79 Chinese articles and 117 English articles) through database searches and finally selected sixteen cohort studies for this meta-analysis. The 16 studies contained a combined total of 517 subjects, and all selected studies reported the mean ADC value (10−3 mm2/s) in DWI in cervical cancer patients treated with RT and CT. Combined standardized mean difference (SMD) suggested that the mean post-RT and mean post-CT ADC values were significantly higher than the mean pre-RT and mean pre-CT ADC values, respectively, in cervical cancer patients (SMD = 2.95, 95% CI = 2.19–3.72, P b 0.001). Ethnicity-stratified analysis revealed that increased ADC values were observed post-RT and post-CT in both Caucasian (SMD = 1.44, 95% CI = 0.93–1.95, P b 0.001) and Asian populations (SMD = 3.32, 95% CI = 2.42–4.22, P b 0.001), compared with the mean ADC values before RT and CT, respectively, in the two subgroups. Further, subgroup analysis based on b-value revealed that higher ADC values were found in cervical cancer patients after RT and CT, compared to before RT and CT treatment, with both b value ≤ 900 (SMD = 3.71, 95% CI = 2.35–5.07, P b 0.001) and N900 (SMD = 2.55, 95% CI = 1.78–3.32, P b 0.001). The mean ADC value in patients without residual tumor post-RT and post-CT treatment was significantly higher than seen in patients with residual tumors (SMD = 0.80, 95% CI = 0.49–1.12, P b 0.001). Conclusion: Our meta-analysis revealed a significant correlation between mean ADC values and the clinical response to RT and CT treatment. Thus, ADC values in DWI may be effective in evaluating the clinical outcome of treatments in cervical cancer patients. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Cervical cancer is the second most common gynecological disease and the third most frequent cause of cancer mortality worldwide [1,2]. Cervical cancer predominantly occurs in women between the third and fifth decade of their life [3]. Approximately, 3 out of 10 cervical cancer patients will have tumor recurrence and succumb to ⁎ Corresponding author at: Department of Oncology, the First Hospital of Qinhuangdao, Wenhua Road No.2, Haigang District, Qinhuangdao 066000, P.R. China. Tel./fax: +86 335 3508121. E-mail address: [email protected] (Z-Z. Fu).
the disease, and the 5-year survival rate of cervical cancers is only 50% [4]. Traditional treatment consists of radical hysterectomy with lymph node dissection, and patients with early stage cervical cancers have a better response with this approach [5]. However, in patients with tumors larger than 4 cm in diameter, the prognosis is poor, and survival rates of 50–60% are seen in patients with large tumors [6]. Such poor results are attributed to increased incidence of risk factors, such as vascular space invasion, lymph node metastases, deep cervical invasion, undiagnosed parametrial extension and radioresistant hypoxic tumor centers [7]. In this context, a recent approach of neoadjuvant therapy, i.e., giving chemotherapy or radiotherapy prior to radical surgery, might shrink tumors, making it easier for surgical
http://dx.doi.org/10.1016/j.mri.2015.02.002 0730-725X/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
2
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
removal and also help destroy tiny tumors or tumor spreads that go undetected [8]. Currently, radiotherapy (RT) and chemotherapy (CT) remain standard therapeutic approaches for cervical cancers and significantly improve survival, and reduce local and distant recurrence [9]. The newer approaches for successful treatment of cancers rely heavily on diagnostic methods to detect cancers early and involve monitoring treatment efficacy. In cervical cancer setting, there is an urgent need for reliable indicators to measure therapeutic response to minimize adverse side effects and to improve quality of life. For this reason, diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) measurement is an attractive diagnostic tool for cervical cancer patients [10,11]. Diffusion-weighted magnetic resonance imaging (DWI) is highly sensitive to cellularity, viscosity and extracellular free water [12,13]. By comparison with other imaging methods, DWI is completely noninvasive and does not involve exposure to ionizing radiation [14]. In addition, DWI does not require contrast medium and permits more frequent examinations, allowing for repeated monitoring throughout the cancer treatment regimen [3]. DWI exploits the physical principle that the degree of water diffusion in biological tissues is determined by tissue cellularity and the integrity of cell membranes [15]. The random Brownian motion of water protons through biological tissues is impeded by the tissue architecture, and normal tissues have low cell numbers and larger extracellular spaces, which allows greater mobility of molecular water. On the other hand, tumors at various stages exhibit increasingly complex architecture,
with densely packed cells, altered membrane permeability and highly restricted extracellular space, all of which impede water mobility. Importantly, ADC value, a quantitative parameter in DWI that measures the rate of water mobility, increases after successful therapy because tumor cell density decreases, resulting in removal of barriers to water motion [16,17]. In clinic, ADC values are vital for planning therapy, in identification of tumor boundaries and in deciding eligibility for surgical resection [18]. Recently, DWI has also been used successfully to evaluate the therapeutic efficacy of non-surgical treatments in breast cancer, bladder cancer and rectal cancer, and avoid radical surgery to prevent unnecessary removal of the affected organ [19–21]. In the context of this study, ADC values in DWI may be helpful in evaluating the efficacy of concurrent chemo-radiotherapy treatment in cervical cancers [11]. Accumulated evidence indicates that DWI can be used as a surrogate biomarker to assess tumor cellularity in cervical cancers [3,22], but contradictory results are also reported [4,23]. Therefore, we analyzed published data by meta-analysis to assess the clinical value of DWI with ADC values in evaluation of response to RT and CT in cervical cancer patients. 2. Materials and methods 2.1. Literature search and data sources Relevant articles were searched using the following bibliographic databases: MEDLINE (Ovid version, New York, 1966 ~ 2013), Science
Fig. 1. Flow chart of literature search and study selection.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
3
Table 1 Main characteristics and methodological quality of eligible studies. First author
Year
Country
Language
Ethnicity
Patients
Age (years)
MRI machine type
b value (s/mm2)
Kuang F. [37] Lu X. H. [32] Gao Y. [29] Cui G. [28] Nakamura K. [2] Makino H. [38] Kim H. S. [22] Hu A. M. [30] Somoye G. [4] Shi Y. B. [36] Zhang Y. [10] Rizzo S. [35] Cao K. [39] Ma W. L. [33] Liu Y. [3] Harry V. N. [17]
2014 2013 2013 2013 2013 2013 2013 2012 2012 2011 2011 2011 2009 2009 2009 2008
China China China China Japan Japan Korea China UK China China Italy China China China UK
English Chinese Chinese Chinese English English English Chinese English Chinese English English Chinese Chinese English English
Asians Asians Asians Asians Asians Asians Asians Asians Caucasians Asians Asians Caucasians Asians Asians Asians Caucasians
75 31 45 45 69 25 24 50 20 18 14 17 21 16 17 20
49.39 (36 ~ 66) 59.4 (39 ~ 73) 43 (26 ~ 71) 49.6 (39 ~ 68) 62 (31 ~ 90) 63.5 (36 ~ 91) 57.3 (35 ~ 76) 56.3 ± 11.8 (30 ~ 76) 45.3 (29.8 ~ 60.8) 46.1 ± 11.3 (31 ~ 71) 44 (26 ~ 71) 50 (35 ~ 75) 48.6 ± 9.3 (33 ~ 70) 52 (43 ~ 62) 49 25 ~ 69) 50 (34 ~ 80)
Siemens 3.0 T GE 1.5 T GE 3.0 T GE 1.5 T Siemens 1.5 T Philips 1.5 T Philips 3.0 T Philips 1.5 T GE 1.5 T Philips 1.5 T GE 1.5 T Siemens 1.5 T GE 1.5 T Siemens 3.0 T GE 1.5 T GE 1.5 T
600 1000 700 700 1000 1000 1000 600 1000 1000 1000 900 1000 1000 1000 1000
ADC = apparent diffusion coefficient.
Citation Index database (Web of Science, Thomson ISI, Philadelphia, PA, USA, 1945 ~ 2013), the Cochrane Library Database (Oxford, UK, Issue 12, 2013), PubMed (National Library of Medicine, Bethesda, MD, USA, 1966 ~ 2013), Embase (Data Star version, Cary, NC, 1974 ~ 2013), CINAHL (CD-ROM Information Retrieval System, 1982 ~ 2013), and Current Contents Index (Institute for Scientific Information, 1995 ~ 2013). We also searched three Chinese databases (Chinese Biomedical
Database, 1978 ~ 2013; the Chinese Journal Full-Text Database, 1980 ~ 2013; and the Weipu Journal Database, 1989 ~ 2013) to identify articles published in Chinese language. We used a combination of the following medical subject and free language terms in a highly sensitive search strategy: (“Diffusion Magnetic Resonance Imaging” or “Diffusion MRI” or “Diffusion Weighted MRI” or “DWI” or “MRI-DWI” or “diffusion-weighted imaging” or “diffusion-weighted-MRI”) and
Fig. 2. NOS scoring of included studies.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
4
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
Fig. 3. Forest plots for the differences in diffusion-weighted magnetic resonance imaging for predicting the efficacy of radiotherapy and chemotherapy for cervical cancer (A: posttreatment vs. pretreatment, B: no residue tumor vs. residue tumor).
(“Chemoradiotherapy” or “radiochemotherapy” or “concurrent chemoradiotherapy” or “synchronous chemoradiotherapy” or “concomitant radiochemotherapy” or “CTRT” or “CT-RT” or “CC-RT”) and (“Uterine Cervical Neoplasms” or “Cervical Neoplasms” or “Cervical cancer” or “cervical carcinoma” or “cervix cancer” or “Cervix Neoplasms” or “Cervix carcinoma”). We combined results of electronic searches with further manual searches to identify other potential articles from cross-references.
2.2. Selection criteria Studies reported in languages unfamiliar to the authors were translated. Full-text of the papers was retrieved if at least one study investigator considered the abstract suitable. Investigators independently assessed the retrieved studies based on the following inclusion criteria: (1) studies must be a clinical cohort study reporting the efficacy of mean ADC values to predict the tumor response to RT and CT treatment of cervical cancer; (2) all patients with pathologically confirmed cervical cancer underwent diffusionweighted magnetic resonance imaging (DW-MRI) before and during RT and CT treatment; (3) the study must contain sufficient information on the mean ADC value for assessing response before and after RT and CT treatment for cervical cancer. Certain publication types, such as abstracts, letters to the editor, and proceedings from scientific meetings and the articles that did not satisfy the current inclusion criteria were excluded. The most recent or the largest
sample size publication was selected when studies using the same subjects were published.
2.3. Data extraction and quality assessment Two investigators independently assessed the data in each study. Disagreements in study selection or data collection were resolved by consensus, and a third reviewer was consulted if necessary. We used a piloted data-extraction sheet, which covered the following descriptive information from included studies: surname and initials of the first author, the year of publication or submission, source country and ethnicity, language of publication, study design, sample size, demographic variables of the subjects, therapy type, MRI type, MRI machine type code, b-value (s/mm 2) and the mean ADC value (10 −3 mm 2/s) before and after RT and CT treatment. The quality of the trials was assessed independently by the investigators using Newcastle-Ottawa Scale (NOS) (http://www.ohri.ca/programs/ clinical_epidemiology/oxford.asp). According to NOS, 9 entries are listed as follows: (1) selection: representativeness of the exposed cohort (NOS01), if exposed cohort and non-exposed cohort come from the same population (NOS02), ascertainment of exposure (NOS03), demonstration that outcome of interest was not present at start of study (NOS04); (2) comparability: control of important factors (NOS05), control of other important mixed factors (NOS06); (3) outcome: independent studies and blinded method (NOS07), was follow-up long enough for outcomes to occur (NOS08), assesses
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
5
A
B
Fig. 4. Subgroup analysis by ethnicity and b value for the differences in diffusion-weighted magnetic resonance imaging for predicting the efficacy of radiotherapy and chemotherapy for cervical cancer (A: ethnicity, B: b value).
the follow-up of the cohorts to ensure that losses are not related to either the exposure or the outcome (NOS9).
2.4. Statistical analysis Standardized mean difference (SMD) was chosen as the summary statistic for meta-analysis and computed with its corresponding 95% confidence interval (95% CI). Pooled estimates of the statistical significance of SMDs were obtained from the Z test. The heterogeneity between trials for each comparison was estimated by use of Cochran's Q-statistic and I 2 tests [24]. If Q-test showed evidence of a P b 0.05 or I 2 test exhibited N 50%, meta-analysis was done using a random-effects model, otherwise fixed-effects model was used [25,26]. Furthermore, univariate and multivariate meta-regression analysis was used to identify heterogeneity, and Monte Carlo simulation was used for multiple testing corrections. A sensitivity analysis was employed to evaluate the impact of single studies on
the overall estimate. To test potential for publication bias across included trials, we created funnel plots and undertook the Egger's linear regression test [27]. All P values were two-tailed, and values P b 0.05 were considered as statistically significant. Statistical analyses were conducted with the STATA statistical software (Version 12.0, Stata Corporation, and College Station, TX, USA). 3. Results 3.1. Description of included studies Fig. 1 shows the flow diagram of study selection process for this meta-analysis. Initially, we retrieved 196 articles by electronic database search. After screening titles and abstracts of all the retrieved articles, we systematically reviewed 101 full text articles, and identified 16 articles based on a careful assessment of quality, relevance and completeness of data [2–4,10,22,23,28–39]. The included studies consisted of a combined total of 517 patients
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
6
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
C
D
E
F
Fig. 5. Meta-regression analysis for the differences in diffusion-weighted magnetic resonance imaging for predicting and assessing cervical cancer (A: publication year, B: sample size, C: b value, D: instrument, E: ethnicity, F: language).
with cervical cancers, with sample size ranging from 28 to 150 between the studies. Of the 16 studies included in the analysis, 13 were conducted in Asians (ten from China, two from Japan, and one from Korea), and the other three were performed in Caucasians (two from UK, and one from Italy). The MRI machine types included with GE 1.5/3.0 T MRI scanners, 1.5 T/3.0 Siemens Avanto scanners, Philips 1.5/3.0 T scanners and Tesla 3.0 T scanners. The mean ADC value before and after RT and CT treatments were obtained. Characteristics of included studies are presented in Table 1, and the NOS assessments for quality are displayed in Fig. 2, which indicated a moderate to high study quality.
Table 2 Meta-regression analyses of potential source of heterogeneity. Heterogeneity factors
Coefficient
Year Sample b value Machine Ethnicity Language
0.194 −0.01 −0.003 0.12 −0.686 1.611
SE
0.339 0.02 0.005 0.275 1.453 1.132
t
0.57 −0.52 −0.68 0.44 −0.47 1.42
P (adjusted)
0.982 0.987 0.959 0.997 0.994 0.555
SE = standard error; LL = lower limit; UL = upper limit.
95% CI LL
UL
−0.572 −0.054 −0.014 −0.501 −3.973 −0.949
0.96 0.034 0.007 0.741 2.601 4.171
3.2. Meta-analysis A total of 16 studies reported differences in the mean ADC values in cervical cancer patients before and after RT and CT treatment. The result of heterogeneity test suggested presence of heterogeneity among the included studies (I 2 = 94.5%, Ph b 0.001), therefore a random-effects model was chosen. The result of meta-analysis showed that the mean post-RT and post-CT ADC values were significantly higher than the mean pre-RT and pre-CT ADC values in cervical cancer patients (SMD = 2.95, 95% CI = 2.19–3.72, P b 0.001) (Fig. 3A). Six studies reported the differences in the mean ADC values in patients with residual tumor and in patients with no residual tumor post-RT and post-CT treatment. Heterogeneity test results suggest that there was no heterogeneity among the included studies (I 2 = 0.0%, Ph = 0.640), thus a fixed-effects model was chosen. Meta-analysis of the results showed that the mean ADC values in patients with no residual tumors was significantly higher than the ADC values observed in patients with residual tumors post-RT and post-CT treatment (SMD = 0.80, 95% CI = 0.49–1.12, P b 0.001) (Fig. 3B). In order to explore potential sources of heterogeneity, we also performed subgroup analyses based on ethnicity and b value. Ethnicitystratified analysis demonstrated that both Caucasians (SMD = 1.44, 95% CI = 0.93–1.95, P b 0.001) and Asians (SMD = 3.32, 95% CI = 2.42–
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
A
B
C
D
7
Fig. 6. Sensitivity analysis and funnel plot of publication biases for the differences in diffusion-weighted magnetic resonance imaging for predicting and assessing cervical cancer (A–B:. sensitivity analysis, C–D: publication bias).
4.22, P b 0.001) (Fig. 4A) showed a similar increase in ADC values post-RT and post-CT, compared with ADC values pre-RT and pre-CT. Further, subgroup analysis based on b-value revealed significantly higher ADC values in cervical cancer patients post-RT and post-CT treatment compared to pre-RT and pre-CT treatment with both b value ≤ 900 (SMD = 3.71, 95% CI = 2.35–5.07, P b 0.001) and N 900 (SMD = 2.55, 95% CI = 1.78–3.32, P b 0.001) (Fig. 4B). Univariate regression analysis showed language as a major source of heterogeneity (P = 0.024), while publication year, sample size, b value, instruments and ethnicity showed no heterogeneity (publication year: P = 0.338; sample size: P = 0.461; b value: P = 0.143; instrument: P = 0.422; ethnicity: P = 0.113) (Fig. 5). Multivariate regression analysis showed that publication year, sample size, b value, instrument, ethnicity, language were not heterogeneity sources (Table 2). 3.3. Sensitive analysis and publication bias Sensitivity analysis showed that all included studies had no effect on the pooled SMD of the mean ADC values in patients with residual tumors and no residual tumors pre- and post-RT and CT treatment (Fig. 6A–B). Funnel plot analysis in relation to the differences of the mean ADC values revealed evidence of obvious asymmetry in cervical cancer patients before and after RT and CT treatment (Fig. 6C). Further Egger's linear regression analysis confirmed the presence of publication bias of the included studies (P = 0.032). However, the funnel plot analysis revealed no evidence of asymmetry in cervical cancer patients with residual tumors and no residual tumors (Fig. 6D), which was further confirmed by Egger's linear regression analysis (P = 0.609).
4. Discussion The present meta-analysis was carried out to assess the efficacy of DWI in evaluating clinical response to RT and CT in cervical cancer patients. ADC measurements are useful for early detection, determination of malignancy and evaluation of treatment response [40]. In general, high ADC values convey that water molecules can move freely in the tissue, suggesting low cellularity and better organization of tissue structure, a typical feature of normal tissues. On the other hand, low ADC values indicate impeded mobility of water molecules, suggesting high cellularity, a typical feature in malignant lesions. Benign lesions, including simple cysts and hemangiomas, also have higher ADC values due to low cell density [41–43]. The major result of our meta-analysis revealed that ADC values of cervical cancer patients post-RT and CT were significant higher than pre-RT and CT, and the ADC values of patients with no residual tumors was significantly higher than patients with residual tumors after RT and CT, indicating that DWI may be effective in evaluating the clinical response to RT and CT. DWI is highly sensitivity to the motion of water molecules and permits non-invasive characterization of biological tissues based water diffusion properties [44,45]. Through the application of diffusion-weighted gradient pulses to conventional magnetic resonance sequence, the sensitivity of the signal to the level of localized water diffusibility, quantified as ADC, can be achieved. ADC obtained from DWI is a novel assessment tool and functions as a biomarker for assessing early response to treatment in various cancer types [23]. ADC can provide useful information on tumor aggressiveness, tumor cellularity, subtype characterization and clinical response to cancer treatment [46]. A significant change
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
8
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx
in ADC values after treatment with RT and CT is regarded as a positive clinical response, while absence of change in ADC values indicates disease progression [47]. Since our results show a significant difference in ADC values pre- and post-RT and CT, we postulate that DWI is effective in evaluating clinical response to RT and CT in cervical cancer patients. Rizzo et al. also observed that ADC values were significantly different between normal and cancer tissue in the cervix, and between cervical tumors before and after nonsurgical therapy in all patients studied [35]. In a study by Makino et al., DWI and ADC measurement were useful in early evaluation of response to RT and CT in cervical cancer patients [34]. Our stratified analysis based on ethnicity and b-value excluded the potential of heterogeneity based on race. In the ethnicitystratified analysis, we found that the ADC values were significantly increased after RT and CT treatment in both Asian and Caucasian populations. Furthermore, the results of subgroup analysis by b-value suggested that there was significant difference in ADC values before and after RT and CT treatment in majority of subgroups, indicating that selection of b-value may not have an impact on the overall results. In summary, our meta-analysis results are consistent with previous studies that showed an elevation in ADC values in cervical cancer patients after RT and CT treatment, supporting the view that DWI is a reliable surrogate biomarker for assessing clinical responses in cervical cancer patients. Our meta-analysis has potential limitations. First, potential sources of bias must be considered: all studies included in our meta-analysis had methodological limitations such as no intentionto-treat analysis, unclear or inadequate data collection. Study limitations also include a small sample size and lack of a uniform standard on ADC values in judging the efficacy of cervical cancer treatment. African populations were not included in this analysis, which may result in publication bias. Second, the studies could be significantly heterogeneous because some of the planned analyses were restricted by small sample size within those trials. A third limitation in our meta-analysis is that we did not consider articles and abstracts. Therefore, the results of this meta-analysis must be interpreted with caution. Despite the above limitations, this is the first example of a meta-analysis to test the association of ADC values in DWI with the clinical response in cervical cancer. In conclusion, our meta-analysis provides evidence that increase in ADC values of DWI is a reliable indicator of favorable treatment response in cervical cancer patients. Nevertheless, due to the limitations of our study, prospective studies are needed in the future to obtain a standardized assessment of diagnostic markers and confirm our findings in a larger patient population. Conflicts of interest No competing financial interests exist. Acknowledgments We would like to acknowledge the reviewers for their helpful comments on this paper. References [1] Bygbjerg IC. Double burden of noncommunicable and infectious diseases in developing countries. Science 2012;337:1499–501. [2] Nakamura K, Kajitani S, Joja I, Haruma T, Fukushima C, Kusumoto T, et al. The posttreatment mean apparent diffusion coefficient of primary tumor is superior to pretreatment ADCmean of primary tumor as a predictor of prognosis with cervical cancer. Cancer Med 2013;2:519–25. [3] Liu Y, Bai R, Sun H, Liu H, Zhao X, Li Y. Diffusion-weighted imaging in predicting and monitoring the response of uterine cervical cancer to combined chemoradiation. Clin Radiol 2009;64:1067–74.
[4] Somoye G, Harry V, Semple S, Plataniotis G, Scott N, Gilbert FJ, et al. Early diffusion weighted magnetic resonance imaging can predict survival in women with locally advanced cancer of the cervix treated with combined chemoradiation. Eur Radiol 2012;22:2319–27. [5] Pfaendler KS, Wenzel L, Mechanic MB, Penner KR. Cervical cancer survivorship: long-term quality of life and social support. Clin Ther 2015;37:39–48. [6] Lintner B, Saso S, Tarnai L, Novak Z, Palfalvi L, Del Priore G, et al. Use of abdominal radical trachelectomy to treat cervical cancer greater than 2 cm in diameter. Int J Gynecol Cancer 2013;23:1065–70. [7] Wang N, Guan QL, Wang K, Zhou X, Gao C, Yang HT, et al. Radiochemotherapy versus radiotherapy in locally advanced cervical cancer: a meta-analysis. Arch Gynecol Obstet 2011;283:103–8. [8] Rydzewska L, Tierney J, Vale CL, Symonds PR. Neoadjuvant chemotherapy plus surgery versus surgery for cervical cancer. Cochrane Database Syst Rev 2012;12:CD007406. [9] Katanyoo K, Sanguanrungsirikul S, Manusirivithaya S. Comparison of treatment outcomes between squamous cell carcinoma and adenocarcinoma in locally advanced cervical cancer. Gynecol Oncol 2012;125:292–6. [10] Zhang Y, Chen JY, Xie CM, Mo YX, Liu XW, Liu Y, et al. Diffusion-weighted magnetic resonance imaging for prediction of response of advanced cervical cancer to chemoradiation. J Comput Assist Tomogr 2011;35:102–7. [11] Levy A, Caramella C, Chargari C, Medjhoul A, Rey A, Zareski E, et al. Accuracy of diffusion-weighted echo-planar MR imaging and ADC mapping in the evaluation of residual cervical carcinoma after radiation therapy. Gynecol Oncol 2011;123: 110–5. [12] Zhang J, Tehrani YM, Wang L, Ishill NM, Schwartz LH, Hricak H. Renal masses: characterization with diffusion-weighted MR imaging–a preliminary experience. Radiology 2008;247:458–64. [13] Kim CK, Park BK, Kim B. Diffusion-weighted MRI at 3 T for the evaluation of prostate cancer. AJR Am J Roentgenol 2010;194:1461–9. [14] Sugita R, Yamazaki T, Furuta A, Itoh K, Fujita N, Takahashi S. High b-value diffusion-weighted MRI for detecting gallbladder carcinoma: preliminary study and results. Eur Radiol 2009;19:1794–8. [15] Razek AA, Megahed AS, Denewer A, Motamed A, Tawfik A, Nada N. Role of diffusion-weighted magnetic resonance imaging in differentiation between the viable and necrotic parts of head and neck tumors. Acta Radiol 2008;49:364–70. [16] Song I, Kim CK, Park BK, Park W. Assessment of response to radiotherapy for prostate cancer: value of diffusion-weighted MRI at 3 T. AJR Am J Roentgenol 2010;194:W477–82. [17] Harry VN. Novel imaging techniques as response biomarkers in cervical cancer. Gynecol Oncol 2010;116:253–61. [18] McVeigh PZ, Syed AM, Milosevic M, Fyles A, Haider MA. Diffusion-weighted MRI in cervical cancer. Eur Radiol 2008;18:1058–64. [19] Park SH, Moon WK, Cho N, Song IC, Chang JM, Park IA, et al. Diffusion-weighted MR imaging: pretreatment prediction of response to neoadjuvant chemotherapy in patients with breast cancer. Radiology 2010;257:56–63. [20] Yoshida S, Koga F, Kobayashi S, Ishii C, Tanaka H, Tanaka H, et al. Role of diffusion-weighted magnetic resonance imaging in predicting sensitivity to chemoradiotherapy in muscle-invasive bladder cancer. Int J Radiat Oncol Biol Phys 2012;83:e21–7. [21] Jung SH, Heo SH, Kim JW, Jeong YY, Shin SS, Soung MG, et al. Predicting response to neoadjuvant chemoradiation therapy in locally advanced rectal cancer: diffusion-weighted 3 Tesla MR imaging. J Magn Reson Imaging 2012;35:110–6. [22] Kim HS, Kim CK, Park BK, Huh SJ, Kim B. Evaluation of therapeutic response to concurrent chemoradiotherapy in patients with cervical cancer using diffusionweighted MR imaging. J Magn Reson Imaging 2013;37:187–93. [23] Harry VN, Semple SI, Gilbert FJ, Parkin DE. Diffusion-weighted magnetic resonance imaging in the early detection of response to chemoradiation in cervical cancer. Gynecol Oncol 2008;111:213–20. [24] Zintzaras E, Ioannidis JP. HEGESMA: genome search meta-analysis and heterogeneity testing. Bioinformatics 2005;21:3672–3. [25] Zintzaras E, Ioannidis JP. Heterogeneity testing in meta-analysis of genome searches. Genet Epidemiol 2005;28:123–37. [26] Zhou L, Yao F, Luan H, Wang Y, Dong X, Zhou W, et al. Three novel functional polymorphisms in the promoter of FGFR2 gene and breast cancer risk: a HuGE review and meta-analysis. Breast Cancer Res Treat 2012;136:885–97. [27] Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Comparison of two methods to detect publication bias in meta-analysis. JAMA 2006;295:676–80. [28] Cui G, Yang LJ, Zhang YP, Guo XF, Wang DK, Shi Y. Diffusion weighted imaging in predicting and assessing therapeutic effect of cervical carcinoma. Chin J Clin (Electron Ed) 2013;7:7573–5. [29] Gao Y, Wang F, Kong LL, Wu LM. Diffusion-weighted imaging in predicting and monitoring the response of advanced cervical cancer to combined radiotherapy. Anhui Med Pharm J 2013;17:967–9. [30] Hu AM, Wang Y, X., Zhou SJ, Zhu ZW, Huang ZH, et al. The applied research of diffusion weighted imaging in cervical carcinoma. J Wenzhou Med Coll 2012;42:474–7. [31] Li YL. The applied research of DWI in monitoring therapeutic effect of uterine cervix cancer with radiotherapy. Xinli Yesheng; 2012 114. [32] Lu XH, Wang J, Liu WY, Ding Y. Diffusion weighted imaging in predicting and assessing therapeutic effect of cervical carcinoma. J Chin Pract Diagn Ther 2013;27:358–60. [33] Ma WL, Huan Y, Wei LC, Lu J, Xu JQ, Zhao HT, et al. The applied research of 3.0 T MRI and DWI in monitoring therapeutic effect of uterine cervix cancer with radiotherapy. J Clin Radiol 2009;28:1651–4. [34] Makino H, Kato H, Furui T, Morishige KI, Kanematsu M. Predictive value of diffusion-weighted magnetic resonance imaging during chemoradiotherapy for uterine cervical cancer. J Obstet Gynaecol Res 2014;40:1098–104.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
Z-Z. Fu et al. / Magnetic Resonance Imaging xxx (2015) xxx–xxx [35] Rizzo S, Summers P, Raimondi S, Belmonte M, Maniglio M, Landoni F, et al. Diffusion-weighted MR imaging in assessing cervical tumour response to nonsurgical therapy. Radiol Med 2011;116:766–80. [36] Shi YB, Tong HZ, Chen MD, Zhou Q. Evaluation of the early chemoradiation efficacy in cervical cancer by diffusion weighted magnetic resonance imaging. J Med Imaging 2011;21:887–9. [37] Kuang F, Yan Z, Wang J, Rao Z. The value of diffusion-weighted MRI to evaluate the response to radiochemotherapy for cervical cancer. Magn Reson Imaging 2014;32: 342–9. [38] Makino H, Kato H, Furui T, Morishige KI, Kanematsu M. Predictive value of diffusion-weighted magnetic resonance imaging during chemoradiotherapy for uterine cervical cancer. J Obstet Gynaecol Res 2013;40:1098–104. [39] Cao K, Zhang XP, Wang N, Shan J, Cui Y, Qi LP. Evaluation of the earl y chemoradiation efficacy in cervical cancer by diffusion-weighted magnetic resonance imaging. Chin J Med Imaging Technol 2009;1657–60. [40] Klimas A, Drzazga Z, Kluczewska E, Hartel M. Regional ADC measurements during normal brain aging in the clinical range of b values: a DWI study. Clin Imaging 2013;37:637–44.
9
[41] Testa ML, Chojniak R, Sene LS, Damascena AS, Guimaraes MD, Szklaruk J, et al. Is DWI/ADC a useful tool in the characterization of focal hepatic lesions suspected of malignancy? PLoS One 2014;9:e101944. [42] Li SP, Padhani AR. Tumor response assessments with diffusion and perfusion MRI. J Magn Reson Imaging 2012;35:745–63. [43] Galea N, Cantisani V, Taouli B. Liver lesion detection and characterization: role of diffusion-weighted imaging. J Magn Reson Imaging 2013;37:1260–76. [44] Padhani AR, Liu G, Koh DM, Chenevert TL, Thoeny HC, Takahara T, et al. Diffusionweighted magnetic resonance imaging as a cancer biomarker: consensus and recommendations. Neoplasia 2009;11:102–25. [45] Hamstra DA, Rehemtulla A, Ross BD. Diffusion magnetic resonance imaging: a biomarker for treatment response in oncology. J Clin Oncol 2007; 25:4104–9. [46] Koh DM, Collins DJ. Diffusion-weighted MRI in the body: applications and challenges in oncology. AJR Am J Roentgenol 2007;188:1622–35. [47] Moffat BA, Chenevert TL, Lawrence TS, Meyer CR, Johnson TD, Dong Q, et al. Functional diffusion map: a noninvasive MRI biomarker for early stratification of clinical brain tumor response. Proc Natl Acad Sci U S A 2005;102:5524–9.
Please cite this article as: Fu Z-Z, et al, Value of apparent diffusion coefficient (ADC) in assessing radiotherapy and chemotherapy success in cervical cancer, Magn Reson Imaging (2015), http://dx.doi.org/10.1016/j.mri.2015.02.002
