PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
 
PLoS One. 2016; 11(6): e0157606.
Published online 2016 June 20. doi:  10.1371/journal.pone.0157606
PMCID: PMC4913930

A Comparative Study of Noninvasive Hypoxia Imaging with 18F-Fluoroerythronitroimidazole and 18F-Fluoromisonidazole PET/CT in Patients with Lung Cancer

Matthew Bogyo, Editor

Abstract

Purpose

This is a clinical study to compare noninvasive hypoxia imaging using 18F-fluoroerythronitroimidazole (18F-FETNIM) and 18F-fluoromisonidazole (18F-FMISO) positron emission tomography/computed tomography (PET/CT) in patients with inoperable stages III–IV lung cancer.

Methods

A total of forty-two patients with inoperable stages III–IV lung cancer underwent 18F-FETNIM PET/CT (n = 18) and 18F-FMISO PET/CT (n = 24) before chemo/radiation therapy. The standard uptake values (SUVs) of malignant and normal tissues depict 18F-FETNIM PET/CT and 18F-FMISO PET/CT uptake. Tumor-to-blood ratios (T/B) were used to quantify hypoxia.

Results

All patients with lung cancer underwent 18F-FETNIM PET/CT and 18F-FMISO PET/CT successfully. Compared to 18F-FMISO, 18F-FETNIM showed similar uptake in muscle, thyroid, spleen, pancreas, heart, lung and different uptake in blood, liver, and kidney. Significantly higher SUV and T/B ratio with 18F-FMISO (2.56±0.77, 1.98±0.54), as compared to 18F-FETNIM (2.12±0.56, 1.42±0.33) were seen in tumor, P = 0.022, <0.001. For the patients with different histopathological subtypes, no significant difference of SUV (or T/B ratio) was observed both in 18F-FMISO and 18F-FETNIM in tumor. A significantly different SUV (or T/B ratio) was detected between < = 2cm, 2~5cm, and >5cm groups in 18F-FMISO PET/CT, P = 0.015 (or P = 0.029), whereas no difference was detected in 18F-FMISO PET/CT, P = 0.446 (or P = 0.707). Both 18F-FETNIM and 18F-FMISO showed significantly higher SUVs (or T/B ratios) in stage IV than stage III, P = 0.021, 0.013 (or P = 0.032, 0.02).

Conclusion

18F-FMISO showed significantly higher uptake than 18F-FETNIM in tumor/non-tumor ratio and might be a better hypoxia tracer in lung cancer.

Introduction

Intratumoral hypoxia increases tumor aggressiveness, chemo- and radio-resistance. Locoregional failure is common in several human cancers, particularly after chemoradiotherapy, and may be attributed to intrinsic tumor resistance to radiotherapy and/or chemotherapy. Hypoxia status has a prognostic value and therapeutic implications [1]. Consequently, multiple methods have been proposed to measure hypoxia, both invasively and noninvasively. Although the invasive polarographic needle electrode method is regarded as the gold standard, this method is invasive, difficult for deep tumor tissues, and represents only a small section of the tumor, often misrepresenting the severity of the tumor hypoxia and resulting in an incorrect prognosis [2]. Recently, PET radiotracers have been intensively investigated as the noninvasive methods for imaging hypoxia within the tumor microenvironment, the common radiotracers including 18F-fluoroerythronitroimidazole (18F-FETNIM) and 18F-fluoromisonidazole (18F-FMISO).

18F-FMISO PET has been widely used to detect tumor hypoxia in clinical [3,4]. Given the pivotal role of tumor hypoxia in cancer treatment response and prognosis, 18F-FMISO PET seems to be a very promising modality in the evaluation of oncological patients. Nevertheless, despite its wide application, 18F-FMISO remains a tracer, whose role in the diagnostic workup of patients with lung cancer is still open. On the other hand, 18F-FETNIM, a radiotracer able to detect hypoxia, has mainly been studied in head and neck cancers, and it seems that high uptake of 18F-FETNIM is associated with a poor outcome in head-and-neck tumors [5]. Recently, in the study of non-small cell lung cancer, 18F-FETNIM tumor/blood ratio (T/Bmax) and hypoxia volume (HV) were strong predictors for overall survival (OS) after treatment, and 18F-FDG uptake of the primary lesions did not have a significant relationship with survival [6]. These original observations prompted us to study the uptake and metabolism of 18F-FMISO and 18F-FETNIM in tumor-bearing mammals [7] and, finally, in cancer patients.

In this pilot study, we compared the noninvasive hypoxia vivo imaging ability of 18F-FMISO PET/CT with that of 18F-FETNIM PET/CT in patients with lung cancer and aimed to find the better noninvasive method for evaluation of hypoxia in lung cancer. A further aim of the study was to assess the biodistribution of the two radiotracers in normal tissues.

Materials and Methods

Ethical approval

The study was approved by the Ethical Committee of the Shandong Cancer Hospital and Institute, and each patient gave written and informed consent before the study.

Patients

Forty-two patients inoperable stage III–IV lung cancer were enrolled in this prospective study before chemo/radiation therapy between February 2014 and August 2015 in Shandong Cancer Hospital. All patients received conventional physical examinations, laboratory tests, and systematically diagnostic staging procedures including enhanced CT, magnetic resonance imaging (MRI), and single photon emission computed tomography (SPECT) bone scan and ultrasonography. The clinical stage was determined by 2010 American Joint Committee on Cancer TNM classification system. The final staging was based on the findings at pathologic examination. They were randomized into two groups, one group of twenty-four patients (twenty-one male, three female) were imaged with 18F-FMISO PET/CT, and another group of eighteen patients (fourteen male, four female) with 18F-FETNIM PET/CT imaging. The study was approved by the Ethical Committee of the Shandong Cancer Hospital and Institute, and each patient gave written and informed consent before the study.

18F-FMISO and 18F-FETNIM PET/CT imaging

18F-FMISO and 18F-FETNIM PET/CT studies were performed before any treatment. All patients underwent PET/CT scan 1 to 7 days before treatment. 18F-FMISO and 18F-FETNIM was prepared as described in previous studies [7,8]. The radiochemical purity of 18F-FMISO and 18F-FETNIM exceeded 95%, and its specific radioactivity exceeded 37 GBq (1,000 mCi)/μmol. No specific subject preparations, patients did not need fasting and did not receive CT contrast agents. Patients received a dose of 4.81-MBq/kg (0.13 mCi/kg) 18F-FMISO and 18F-FETNIM intravenously and then rested for approximately 120 minutes [9]. Scanning of chest was performed with an integrated PET/CT device (Discovery LS; GE Healthcare). The spiral CT component was performed with an x-ray tube voltage peak of 140 kV and 80 mA, a 6:1 pitch, a slice thickness of 4.25 mm, and a rotation speed of 0.8 s per rotation. A full-ring dedicated PET scan of the same axial range followed. The patients were in normal shallow respiration during image acquisition. The images were attenuation corrected with the transmission data from CT. The attenuation-corrected PET images, CT images, and fused PET/CT images displayed as coronal, sagittal, and transaxial slices were viewed on a Xeleris workstation (GEHealthcare).

Image analysis

Two experienced nuclear medicine physicians read all of the images through consensus reading. They were blinded to the clinical and structural imaging findings. PET data were reconstructed using the ordered-subsets expectation maximization algorithm. The SUV was calculated according to the following formula: [measured activity concentration (Bq/mL) × body weight (g)] / injected activity (Bq). The frame on CT images of PET/CT was used to define the whole primary tumor. Regions of interest with a diameter of 1.5 cm were drawn on the pulmonary artery (for measurement of blood activity), heart, lung, liver, thyroid gland, spleen, bone, and muscle with the assistance of corresponding CT images. Because of the heterogeneity of tumor tissue, a maximum area of 3 × 3 pixels (7.04 × 7.04 mm) inside each tumor region was also determined, with an automated system to represent the highest radioactivity concentration in the tumor. The results were expressed in mean SUV. Finally, the tumor-to-blood (T/B) ratio, the ratio of the peak SUV of the tumor to the SUV of the pulmonary artery, was determined.

Measurement of tumor size

All the patients measured the longest diameter of the primary tumor on CT images by two experienced nuclear medicine physicians. Patients were divided into 3 groups (< = 2cm; >2cm, < = 5cm; and >5cm) based on the longest diameter of tumor and further to investigate the correlation between different tumor size and hypoxia.

Statistics

All data are presented as mean±SD unless otherwise stated. Statistical analysis was performed with the SPSS, version 17.0. The non-parametric Mann-Whitney U test was used to evaluate the significance of differences in the uptake of the markers between the control patients. Differences between continuous variables and dichotomous variables were tested by non-parametric Kruskal-Wallis H test. A P value of less than 0.05 was considered statistically significant.

Results

Patient characteristics and safety

The patient characteristics were summarized in Table 1. There was 1 patient missing the tumor diameter measurement because of obstructive pneumonia in the 18F-FETNIM study, and 1 patient with obstructive atelectasis in the 18F-FMISO study. There was no statistical difference in age, histopathology types, tumor size and staging between the two groups of patients, P = 0.452, 0.775, 0.471, 0.341.

Table 1
Demographic profile and histological diagnosis in 42 patients with lung cancer.

After the examination, there were no adverse or clinically detectable pharmacologic effects in any of the subjects. No significant changes in vital signs or the results of laboratory studies or electrocardiograms were observed.

Biodistribution

The biodistribution of 18F-FETNIM and 18F-FMISO in the normal tissues is shown in Table 2. No significantly different uptake with 18F-FETNIM as compared to18F-FMISO was seen in muscle, thyroid, spleen, pancreas, heart, and lung. The highest accumulation activity was found in the kidneys, demonstrating renal clearance. The liver also showed moderate uptake. A significantly higher uptake with 18F-FETNIM as compared to18F-FETNIM was seen in blood, liver, and kidney, P = 0.007, <0.001, <0.001.

Table 2
Biodistribution of 18F-FETNIM and 18F-FMISO on PET/CT in Patients with Lung Cancer.

Hypoxia detection for tumor

The result of the SUV and T/B ratio measurement for tumors assessed by 18F-FETNIM and 18F-FMISO PET CT was shown in Table 2 and Fig 1. Significantly higher SUV and T/B ratio with 18F-FMISO (2.56±0.77, 1.98±0.54) as compared to 18F-FETNIM (2.12±0.56, 1.42±0.33), were seen in tumor, P = 0.022, <0.001.

Fig 1
PET imaging.

Semiquantitative assessment of 18F-FETNIM and 18F-FMISO uptake in tumor under different conditions was presented in Table 3. No significant difference of SUV (or T/B ratio) was observed between different histopathological subtypes both in 18F-FETNIM and 18F-FMISO in tumor, P = 0.121, 0.483 (or P = 0.221, 0.362). For 18F-FETNIM PET/CT, there was no correlation between the tumor size and the SUV (or T/B ratio) in tumors (P = 0.446, P = 0.707). With18F-FMISO, a significant increase in SUV (or T/B ratio) was seen along with the increase of tumor longest diameter (P = 0.015, P = 0.029). The SUV and T/B ratio were significantly higher in stage IV tumor than in stage III tumor for both 18F-FETNIM and 18F-FMISO, P = 0.021, 0.032, 0.013, 0.02.

Table 3
Semiquantitative Assessment of 18F-FETNIM and 18F-FMISO Uptake in Tumor Under Different Conditions.

Discussion

42 patients with lung cancer were enrolled in the hypoxia imaging prospective study and our results indicated that 18F-FMISO PET/CT may play a better role in hypoxia detection in patients with lung cancer than that of 18F-FETNIM PET/CT. For clinical use of the hypoxia radiotracers, the PET signal needs to be quantified. Both for 18F-FMISO and 18F-FETNIM PET/CT considered T/B ratio threshold of a fixed value as a parameter to discriminate hypoxic from normoxic tissues [6,10].

In the biodistribution studies, there was no significantly different uptake between 18F-FETNIM and18F-FMISO on PET/CT in most of the normal tissues except for blood, kidneys, and liver. Hypoxia in tumors is a pathophysiologic consequence of structurally and functionally disturbed angiogenesis along with deterioration in the inability of oxygen to diffuse through tissues [11]. PET has the ability to differentiate individual tumor biology more accurately and modify treatment accordingly and the tracers targeting tissue hypoxia provides further tools to evaluate the molecular profile of tumors [12]. In this study, we validated the use of PET/CT in detection tumor hypoxia under different histopathological subtypes, stages, and tumor size with 18F-FMISO and 18F-FETNIM in patients with lung cancer.

In the current study, there was no difference between different histopathological subtypes with 18F-FMISO or 18F-FETNIM uptake in patients with lung cancer. For the 18F-FMISO study, the uptake in gliomas in correlation with histological findings was previously investigated by Cheret al [13]. Yamamoto et al had also found that there was a correlation between 18F-FMISO uptake and glioma grade, and T/B ratio in grade IV gliomas was significantly higher than all other grades [14]. Although 18F-FETNIM has been developed to evaluated hypoxia in human [15,16] tumors since 1995 [17], to our best knowledge, there are few reports about correlation between hypoxia and pathological subtypes in lung cancer.

18F-FMISO showed a potential advantage to discriminate the heterogeneous of oxygenation within the tumor and it also showed a significantly different T/B ratio or SUV than 18F-FETNIM in different tumor size in patients with lung cancer. The relationship between 18F-labelled hypoxia marker uptake and tumor size has been investigated in several preclinical studies [18,19]. Different from the current study, no significant correlation was reported between tumor size and uptake ratio in experimental glioma [18] or murine mammary carcinoma [19], when measured with 18F-FMISO. Cells become hypoxic when they exist at distances greater than 1–2 mm from vasculature, where diffusion of oxygen is no longer possible [20]. In theory, 18F-labelled hypoxia marker uptake will depend on size, vascularity, and developmental stage of the xenograph. Xenograft heterogeneity in mice model is smaller than that of patient, and that may explained why the tumor size has no correlation between tumor hypoxia. On the other hand, Chung et al [21] reported a moderate correlation between tumor weight and the tumor uptake of 18F-FETNIM in a murine sarcoma model. However, no different 18F-FETNIM uptake was observed between different tumor sizes in patient with lung cancer in this study.

We also analyzed the correlation between 18F-FMISO or 18F-FETNIM uptake and different stages in patient with inoperable stages III–IV lung cancer, and we found that T/B ratio or SUVmax in stage IV was significantly higher than stage III. As we all know, patient with stage IV lung cancer has a poor prognosis than stage III as hypoxia in tumors is associated with propagation and progression, as well as resistance to radiotherapy and some forms of chemotherapy [2224]. Recent clinical studies have indicated that high 18F-FETNIM SUV may be predictive of treatment outcome in non–small cell lung cancer and esophageal squamous cell tumors [6,25]. The ability to determine the degree and extent of tumor hypoxia is important both as a prognostic biomarker [26] and as a means of selection of patients for hypoxia-directed therapies. The intratumor spatial distribution of hypoxia can potentially serve as a target in radiotherapy planning [27] and also indicate the use of a growing number of adjunct therapies [28]. In this study we further to validate the hypoxia detection ability of 18F-FMISO and 18F-FETNIM in patient with lung cancer.

Although this prospective study was performed successfully, it has the following deficiencies. The major drawback of the current study is that we have no definitive proof of the presence of hypoxia in the tumors of our series and larger cohorts are required to validate these findings in our study.

Conclusion

In this study, we found that the uptake with18F-FMISO was significantly higher than 18F-FETNIM in tumors and might be a better hypoxia tracer for the patients with lung cancer.

Supporting Information

S1 File

Patient Characteristics and Imaging Data of 18F-FETNIM PET/CT.

(DOCX)

S2 File

Patient Characteristics and Imaging Data of 18F-FMISO PET/CT.

(DOCX)

Acknowledgments

The authors would like to thank Ms. Feng-Ming Kong, Mr. Xuepeng Teng, Mr. Kai Cheng, Mr. Zhiguo Liu, Mr. Jianzhong Chen, Ms. Fenghuan Niu, Ms. Hongbo Wu, Mr. Fu Zheng, Mr. Zongwei Huo, Ms. Guifang Zhang, Ms. Li Ma, Mr. Yongchun Cui, Ms. Min Liu, and Ms. Yu Han for excellent technical assistance.

Funding Statement

This study was partly supported by several grants from Natural Science Foundation of China (NSFC81172133, NSFC81372413), the special fund for Scientific Research in the Public Interest (201402011), and the Outstanding Youth Natural Science Foundation of Shandong Province (JQ201423). No other potential conflicts of interest relevant to this article are reported.

Data Availability

Data Availability

All relevantin data are within the paper and its Supporting Information files.

References

1. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res. 1996; 56: 4509–4515. [PubMed]
2. Ceyssens S, Van Laere K, de Groot T, Goffin J, Bormans G, Mortelmans L. [11C]methionine PET, histopathology, and survival in primary brain tumors and recurrence. AJNR Am J Neuroradiol. 2006; 27: 1432–1437. [PubMed]
3. Askoxylakis V, Dinkel J, Eichinger M, Stieltjes B, Sommer G, Strauss LG, et al. Multimodal hypoxia imaging and intensity modulated radiation therapy for unresectable non-small-cell lung cancer: the HIL trial. Radiat Oncol. 2012; 7: 157 doi: 10.1186/1748-717X-7-157 [PMC free article] [PubMed]
4. Rajendran JG, Schwartz DL, O'Sullivan J, Peterson LM, Ng P, Scharnhorst J, et al. Tumor hypoxia imaging with [F-18] fluoromisonidazole positron emission tomography in head and neck cancer. Clin Cancer Res. 2006; 12: 5435–5441. doi: 10.1158/1078-0432.CCR-05-1773 [PMC free article] [PubMed]
5. Lehtio K, Eskola O, Viljanen T, Oikonen V, Gronroos T, Sillanmaki L, et al. Imaging perfusion and hypoxia with PET to predict radiotherapy response in head-and-neck cancer. Int J Radiat Oncol Biol Phys. 2004; 59: 971–982. doi: 10.1016/j.ijrobp.2003.12.014 [PubMed]
6. Li L, Hu M, Zhu H, Zhao W, Yang G, Yu J. Comparison of 18F-Fluoroerythronitroimidazole and 18F-fluorodeoxyglucose positron emission tomography and prognostic value in locally advanced non-small-cell lung cancer. Clin Lung Cancer. 2010; 11: 335–340. doi: 10.3816/CLC.2010.n.042 [PubMed]
7. Gronroos T, Eskola O, Lehtio K, Minn H, Marjamaki P, Bergman J, et al. Pharmacokinetics of [18F]FETNIM: a potential marker for PET. J Nucl Med. 2001; 42: 1397–1404. [PubMed]
8. Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol. 2010; 17: 1471–1474. doi: 10.1245/s10434-010-0985-4 [PubMed]
9. Gronroos T, Bentzen L, Marjamaki P, Murata R, Horsman MR, Keiding S, et al. Comparison of the biodistribution of two hypoxia markers [18F]FETNIM and [18F]FMISO in an experimental mammary carcinoma. Eur J Nucl Med Mol Imaging. 2004; 31: 513–520. doi: 10.1007/s00259-003-1404-x [PubMed]
10. Valable S, Petit E, Roussel S, Marteau L, Toutain J, Divoux D, et al. Complementary information from magnetic resonance imaging and (18)F-fluoromisonidazole positron emission tomography in the assessment of the response to an antiangiogenic treatment in a rat brain tumor model. Nucl Med Biol. 2011; 38: 781–793. doi: 10.1016/j.nucmedbio.2011.01.010 [PubMed]
11. Dachs GU, Tozer GM. Hypoxia modulated gene expression: angiogenesis, metastasis and therapeutic exploitation. Eur J Cancer. 2000; 36: 1649–1660. [PubMed]
12. Phelps ME. PET: the merging of biology and imaging into molecular imaging. J Nucl Med. 2000; 41: 661–681. [PubMed]
13. Cher LM, Murone C, Lawrentschuk N, Ramdave S, Papenfuss A, Hannah A, et al. Correlation of hypoxic cell fraction and angiogenesis with glucose metabolic rate in gliomas using 18F-fluoromisonidazole, 18F-FDG PET, and immunohistochemical studies. J Nucl Med. 2006; 47: 410–418. [PubMed]
14. Yamamoto Y, Maeda Y, Kawai N, Kudomi N, Aga F, Ono Y, et al. Hypoxia assessed by 18F-fluoromisonidazole positron emission tomography in newly diagnosed gliomas. Nucl Med Commun. 2012; 33: 621–625. doi: 10.1097/MNM.0b013e3283529984 [PubMed]
15. Lehti? K, Oikonen V, Gr?nroos T, Eskola O, Kalliokoski K, Bergman J, et al. Imaging of blood flow and hypoxia in head and neck cancer: initial evaluation with [(15)O]H(2)O and [(18)F]fluoroerythronitroimidazole PET. J Nucl Med. 2001; 42: 1643–1652. [PubMed]
16. Lehti? K, Oikonen V, Nyman S, Gr?nroos T, Roivainen A, Eskola O, et al. Quantifying tumour hypoxia with fluorine-18 fluoroerythronitroimidazole ([18F]FETNIM) and PET using the tumour to plasma ratio. Eur J Nucl Med Mol Imaging. 2003; 30: 101–108. doi: 10.1007/s00259-002-1016-x [PubMed]
17. Yang DJ, Wallace S, Cherif A, Li C, Gretzer MB, Kim EE, et al. Development of F-18-labeled fluoroerythronitroimidazole as a PET agent for imaging tumor hypoxia. Radiology. 1995; 194: 795–800. doi: 10.1148/radiology.194.3.7862981 [PubMed]
18. Tochon-Danguy HJ, Sachinidis JI, Chan F, Chan JG, Hall C, Cher L, et al. Imaging and quantitation of the hypoxic cell fraction of viable tumor in an animal model of intracerebral high grade glioma using [18F]fluoromisonidazole (FMISO). Nucl Med Biol. 2002; 29: 191–197. [PubMed]
19. Bentzen L, Keiding S, Horsman MR, Gr?nroos T, Hansen SB, Overgaard J. Assessment of hypoxia in experimental mice tumours by [18F]fluoromisonidazole PET and pO2 electrode measurements. Influence of tumour volume and carbogen breathing. Acta Oncol. 2002; 41: 304–312. [PubMed]
20. Folkman J. What is the evidence that tumors are angiogenesis dependent. J Natl Cancer Inst. 1990; 82: 4–6. [PubMed]
21. Chung JK, Chang YS, Lee YJ, Kim YJ, Jeong JM, Lee DS, et al. The effect of tumor size on F-18-labeled fluorodeoxyglucose and fluoroerythronitroimidazole uptake in a murine sarcoma model. Ann Nucl Med. 1999; 13: 303–308. [PubMed]
22. Brown JM. Therapeutic targets in radiotherapy. Int J Radiat Oncol Biol Phys. 2001; 49: 319–326. [PubMed]
23. Tsang RW, Fyles AW, Milosevic M, Syed A, Pintilie M, Levin W, et al. Interrelationship of proliferation and hypoxia in carcinoma of the cervix. Int J Radiat Oncol Biol Phys. 2000; 46: 95–99. [PubMed]
24. Brizel DM, Dodge RK, Clough RW, Dewhirst MW. Oxygenation of head and neck cancer: changes during radiotherapy and impact on treatment outcome. Radiother Oncol. 1999; 53: 113–117. [PubMed]
25. Yue J, Yang Y, Cabrera AR, Sun X, Zhao S, Xie P, et al. (2012) Measuring tumor hypoxia with 1?F-FETNIM PET in esophageal squamous cell carcinoma: a pilot clinical study. Dis Esophagus 25: 54–61. doi: 10.1111/j.1442-2050.2011.01209.x [PubMed]
26. Vaupel P, H?ckel M, Mayer A. Detection and characterization of tumor hypoxia using pO2 histography. Antioxid Redox Signal. 2007; 9: 1221–1235. doi: 10.1089/ars.2007.1628 [PubMed]
27. Ling CC, Humm J, Larson S, Amols H, Fuks Z, Leibel S, et al. Towards multidimensional radiotherapy (MD-CRT): biological imaging and biological conformality. Int J Radiat Oncol Biol Phys. 2000; 47: 551–560. [PubMed]
28. Harada H. How can we overcome tumor hypoxia in radiation therapy. J Radiat Res. 2011; 52: 545–556. [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science