This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Presant, C. A. Articles by Shani, J. Search for Related Content PubMed PubMed Citation Articles by Presant, C. A. Articles by Shani, J.

Enhancement of Fluorouracil Uptake in Human Colorectal and Gastric Cancers by Interferon or by High-Dose Methotrexate: An In Vivo Human Study Using Noninvasive ¹⁹F-Magnetic Resonance Spectroscopy

From the Los Angeles Oncologic Institute at the St. Vincent Medical Center; University of Southern California School of Pharmacy; and California Cancer Medical Center, Los Angeles, CA; and the David R. Bloom Center for Pharmacy at the Hebrew University of Jerusalem, Jerusalem, Israel.

Address reprint requests to Walter Wolf, PhD, Pharmacokinetic Imaging Program, University of Southern California, 1985 Zonal Ave, Los Angeles, CA 90033; email wwolfw{at}hsc.usc.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To study whether two modulators, high-dose methotrexate (MTX) and interferon alfa-2a (IFN-2a) will alter the intratumoral pharmacokinetics of fluorouracil (5-FU).

PATIENTS AND METHODS: Five patients, two with gastric cancer and three with colorectal cancer, who had metastatic tumor nodules in their livers were studied dynamically in vivo after 5-FU injection. In a magnetic resonance imaging unit, noninvasive ¹⁹F-magnetic resonance spectroscopy (MRS) was used to detect ¹⁹F signals from 5-FU and its metabolites.

RESULTS: The intratumoral half-life (t_1/2) of 5-FU in these tumors ranged from 18.8 minutes to 42.3 minutes. Four of the five patients exhibited increases in the t_1/2 of 5-FU after intravenous (IV) administration of MTX or IFN-2a. In the two patients with gastric cancer who received IV high-dose MTX followed by IV 5-FU, increases were seen in either the total t_1/2 of 5-FU (41.8%) or in the t_1/2 of the phase (150%). In the three patients with colorectal cancer who received IV IFN-2a followed by IV 5-FU, the two patients with partial responses had increases in the t_1/2 of 5-FU of 41% and 30.2%, whereas the nonresponder had a nonsignificant increase (5.6%) in the t_1/2 of 5-FU.

CONCLUSIONS: These results document that the in vivo modulation of the tumoral pharmacokinetics of 5-FU can be measured noninvasively by ¹⁹F-MRS and suggest that such information correlates with subsequent clinical outcomes. The findings also indicate that IFN-2a and high-dose MTX can increase the intratumoral 5-FU in some patients. Such information, obtained prospectively in vivo, may assist in better individual cancer patient management and in developing novel drug combinations.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
FLUOROURACIL (5-FU) has been used for more than 30 years for treating cancers of the gastrointestinal tract, breast, head and neck, ovary, and liver. The efficacy of treatments with 5-FU as a single agent has been considered to be less than 30%.¹ The major mechanisms of 5-FU’s antitumor effect are (1) its inhibition of thymidine synthase (TS), an enzyme that catalyzes the conversion of deoxyuridine monophosphate to thymidylate monophosphate, by fluorodeoxyuridylate (FdUMP); (2) alteration of RNA functionality by fluorouridine triphosphate incorporation; (3) and incorporation of fluorodeoxyuridine triphosphate into DNA.¹

In earlier studies from this laboratory, it was demonstrated that intratumoral pharmacokinetics of 5-FU could be reproducibly measured in vivo in various human cancers.^2,3 In some patients, the intratumoral half-life (t_1/2) of 5-FU was longer than that of other patients with the same type of tumor^2,3; ie, 5-FU was "trapped" in certain tumors. We suggested that this phenomenon was related to differences in clinical responsiveness between patients to this drug. We also demonstrated a highly significant association between the trapping of 5-FU by a tumor of a specific patient and that patient’s clinical response.⁴ A confirmatory study is under way in the Southwest Oncology Group (SWOG 9006). These findings indicate that ¹⁹F-magnetic resonance spectroscopy (MRS) may be a useful methodology to evaluate and predict patient responsiveness to 5-FU chemotherapy.

A variety of drugs have been tested as modulators for 5-FU to increase its chemotherapeutic efficacy. Such modulators have been the basis for the numerous combination protocols for this drug. The modulators that have been used most widely in clinical practice are folinic acid (leucovorin)⁵ and methotrexate (MTX),⁶ along with trimetrexate,⁷ interferon (IFN),⁸ and many others. In the present study, we wish to report that the noninvasive methodology of ¹⁹F-MRS can be used to study the effect of such modulators, in human tumor in vivo, on the t_1/2 of 5-FU in individual patients.

IFN alfa-2a has been known to potentiate the cytotoxicity of 5-FU in vitro and in vivo because of increased conversion of 5-FU to its metabolite FdUMP. IFN-2a also affects DNA repair and the transcriptional upregulation of the thymidylate synthase gene.⁹ IFN has also been demonstrated to potentiate the activity of 5-FU in colon carcinoma in animal models and in the clinic.^10,11 Studies in human colon carcinoma cells have demonstrated that the metabolic activation of 5-FU by IFN-2a seems to be caused by an IFN-induced increase in the thymidine phosphorylation activity, the first enzyme in the pathway of the direct conversion of 5-FU to deoxyribonucleotides. These biochemical events are further enhanced by increased cellular formation of FdUMP.¹² Because 5-FU remains the only cytotoxic agent for the treatment of advanced colon cancer, its use with the biochemical modulator IFN-2a has been studied with the aim of overcoming tumor resistance to 5-FU. The combination of 5-FU and IFN-2a has demonstrated activity against metastatic colorectal cancer in several phase II studies, with a response rate between 26% and 63%. In a multicenter, multinational phase III study with colorectal cancer, there was an equal response rate, response duration, and survival for 5-FU + IFN-2a compared with 5-FU + leucovorin.¹³ In another phase III trial comparing the efficacy of 5-FU with or without IFN-2a for advanced colorectal cancer, the superiority of 5-FU + IFN-2a combination over 5-FU alone was highly significant.¹⁴

MTX has only limited activity by itself in the treatment of advanced colorectal cancer, but it significantly potentiates 5-FU cytotoxicity when both compounds are administered in sequence, with 5-FU administered a few hours after MTX. Sequence-dependent potentiation of 5-FU and MTX has been demonstrated in patients with advanced colorectal carcinoma.¹⁵ The most widely accepted mechanism of action proposed for this modulation is that, by elevating the intracellular amounts of phosphoribosyl-pyrophosphate,¹⁶ there is an increase in the level of 5-FU nucleotides, especially fluorouridine monophosphate and FdUMP, leading to alteration in RNA synthesis and to increased inhibition of TS. Another mechanism of action that has been proposed for this modulation is that the accumulation of dihydrofolate and dihydrofolate polyglutamate results in inhibition of TS.¹⁷ Finally, from work in our laboratory,¹⁸ we have postulated that MTX also seems to have a significant effect on the transport of 5-FU into tumor cells. When 5-FU was administered 5 hours after MTX to rats bearing the Walker 256 adenocarcinoma, the rate constant for the conversion of 5-FU to fluorinated nucleosides and nucleotides had increased by 2.5-fold, whereas that of the disappearance of free 5-FU from the tumor had decreased by three orders of magnitude.

In a later study by Katzir et al,¹⁹ different MTX pretreatment schedules were investigated in the same tumor model for their enhancements of 5-FU activity. Formation of FNUC was enhanced in all experimental schedules, and the amount of FNUC gradually increased as the pretreatment interval expanded up to 4 hours but decreased afterward that time. It was also discovered that the change in the fluorinated product in the RNA follows a pattern similar to that of the FNUC.

We decided, therefore, to use the ¹⁹F-MRS methodology for evaluating in vivo the effect of two modulators, IFN-2a and MTX (in a high dose), on the retention of 5-FU by human tumors (colorectal carcinoma for the 5-FU + IFN-2a protocol and gastric carcinoma for the 5-FU + MTX protocol). The latter study was performed as a single institution study for the Southwest Oncology Group (SWOG 9118).

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient Inclusion and Exclusion Criteria
All patients signed voluntary informed consents for these studies approved by the institutional review board. Patients were eligible to participate in this study regardless of how much prior chemotherapy they may have had, unless they had been treated previously with IFN-2a or MTX. Patients had to have a Karnofsky performance status of more than 70% and were only eligible after recovering from all prior chemotherapeutic side effects. Inclusion criteria required a tumor that was able to be imaged, was not more than 8 cm from the skin (allowing for placement of a magnetic resonance surface coil), and was at least 2 cm in diameter. Patients with pacemakers, metallic implants, or severe claustrophobia were also excluded. Patients with colorectal cancer were treated with 5-FU and IFN-2a. Patients with gastric cancer were treated with 5-FU and MTX.

MRS Studies
Two sequential ¹⁹F-MRS studies were performed on each patient. The first study was performed after an intravenous dose of 5-FU of 600 mg/m² (for the colorectal cancer patients) or 1,500 mg/m² (for the gastric cancer patients). The second ¹⁹F-MRS study was performed 2 to 5 weeks later. The three colorectal patients were given IFN-2a 10 IU/m² subcutaneously 3 times a week and, after 2 weeks of IFN-2a, another dose of 5-FU 600 mg/m². The two gastric cancer patients received a high-dose MTX (1,500 mg/m²), followed 1 hour later by 5-FU 1,500 mg/m² and then oral leucovorin. After these MRS studies, the patients continued to be treated clinically until tumor progression.

Data Acquisition
First, patients were positioned in the magnet, a Magnetom-Helicon SP (Siemens Medical Systems, Iselin, NJ) operating at 1.5 T. Then, the surface coil most suitable for the patient’s tumor was selected, proper shimming occurred, and a background spectrum was acquired. The surface coil, able to both transmit and receive, was tuned to the ¹⁹F frequency. An external reference standard of 1,2-difluorobenzene was incorporated into the surface coil. The dose of 5-FU was administered as an intravenous bolus. Blocks of 256 free induction decays (FIDs) were acquired from the moment of injection for a total of 4.17 minutes. The sequence characteristics were repetition time = 1,000 msec, pulse width ± 2000 Hz, delay time = 250 µsec, vector size 512, with no prescans. An adiabatic half-passage radiofrequency pulse²⁰ was used in all data acquisitions. Serial spectra were collected for up to approximately 80 minutes. Both unlocalized and 1-dimensional chemical shift imaging data were acquired in an interleaved manner. The unlocalized spectral data collected all the ¹⁹F signals detected by the surface coil. These were the data used in the kinetic analyses discussed below. An exponential filter matched to the T2* of the data was used for apodization of the FIDs. The spectral data were also thoroughly checked for the possible presence of peaks of the fluoro-beta-alanine catabolite and of any of the anabolites (FNUC).

Data Processing
The peak intensities and the areas of the real part of 5-FU and of the 1,2-difluorobenzene reference standard were measured after phasing and used to estimate the t_1/2 of the 5-FU peak, using a suitable data analysis program such as Kaleidagraph (Synergy Software, Reading, PA). For compartment modeling, analysis, and simulation the software used was ADAPT-II (Biomedical Simulations Resource, University of Southern California, Los Angeles, CA).

Clinical Evaluation
Patients were evaluated for antitumor response on a monthly basis. Tumor response was measured according to cooperative group criteria.²¹ Patients were considered to have a partial response if the sum of the products of the cross-sectional diameters of their measured tumor areas (on their clinical computed tomography studies) had decreased by more than 50%. The duration of remission was calculated from the first observation of criteria that satisfied the definition of partial response until the time of progression. Survival was measured from the date of study until the time of death.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Five patients were treated in this study. The demographics are listed in Table 1. All five patients received prior chemotherapy (5-FU + leucovorin) and had shown no response. Their age span was 49 to 80 years, and the tumor sites studied were liver metastases. The control baseline 5-FU studies without modulators showed t_1/2s of intratumoral 5-FU ranging from 18.8 minutes to 42.3 minutes. Four of the five patients had long t_1/2s (> 20 minutes), and, of these patients, those who received the 600-mg/m² test dose of 5-FU were classified as trappers by our prior criteria.³

View this table: [in this window] [in a new window]   Table 1. Comparative Tumoral Pharmacokinetic Parameters of Five Cancer Patients Treated With Either 5-FU + IFN-2a or with 5-FU + High-Dose MTX

Out of the three patients with colorectal cancer, patient no. 57, who had failed prior treatment with 5-FU + leucovorin, had an initial t_1/2 of 5-FU of less than 20 minutes and was accordingly classified as a nontrapper. She became a 5-FU trapper (t_1/2 = 26.5 minutes) after 5-FU + IFN-2a. She experienced a 41% increase in the t_1/2 of 5-FU and a partial response of her tumor to 5-FU + IFN-2a. The duration of response was 5 months, and her survival time was 11 months. Patient no. 68, whose initial t_1/2 of 5-FU was more than 20 minutes and who was a trapper, had a smaller increase in the tumoral t_1/2 of 5-FU after modulation with IFN-2a (30%). He experienced a partial response to 5-FU + IFN-2a. The response duration was 4 months, and survival was 6 months. Patient no. 73 exhibited no increase in the t_1/2 of 5-FU and no clinical response and survived for 4 months.

Graphic presentations of the levels of 5-FU seen by the coil before and after the modulator in these three patients are presented in Fig 1 A, 1B, and 1C. These graphs illustrate the effect of the modulator on the kinetics of 5-FU.

View larger version (15K): [in this window] [in a new window]   Fig 1. (A) 5-FU before and after IFN-2a in patient no. 57, analyzed as a one-compartment model. (B) 5-FU before and after IFN-2a in patient no. 68, analyzed as a one-compartment model. (C) 5-FU before and after IFN-2a in patient no. 73, analyzed as a one-compartment model.

View larger version (16K): [in this window] [in a new window]   Fig 2. (A) 5-FU before and after MTX in patient no. 42, analyzed as a two-compartment model. (B) 5-FU before and after MTX in patient no. 60, analyzed as a one-compartment model.

View larger version (23K): [in this window] [in a new window]   Fig 3. Spectra of 5-FU before and after MTX administration to patient no. 42. Note that the intensity of the FNUC peak at 6.25 pulses per minute is approximately 5% of that of 5-FU at 1.24 pulses per minute.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

Although we had anticipated, in analogy to the prior animal study,¹³ that these increased t_1/2s of 5-FU would be accompanied by an increase in the levels of detectable intratumoral anabolites, observable FNUC peaks could be observed in only one of the two patients (Fig 1C) who received high-dose 5-FU modulated by high-dose MTX and in none of the patients who received 5-FU modulated by IFN-2a. It should be noted that although MRS is a technique that provides exquisite chemical information, it also has limited sensitivity, and concentrations of fluoropyrimidines of at least 0.3 mmol are required in the tumor for them to be detectable under the conditions (1.5 T, 4.17 minutes per 256 FIDs) of the present studies. We have not observed signals in the FNUC region in any of the patients studied to date who received only bolus 5-FU 600 mg/m². Further technical improvements, including the possible clinical use of higher magnetic fields (3 or 4T) or more sensitive detector coils, may be needed to allow detection of fluoropyrimidines present at lower concentrations in the tumors and other tissues of patients.

The transfer of 5-FU into tumor cells is represented by the flow-chart and the compartmental model shown in Fig 4. 5-FU transfers from (and back to) the tumoral blood compartment to the interstitial fluid compartment, and from there to the tumor cytoplasm. Given the rapid elimination of 5-FU from the blood, the 5-FU detected by noninvasive MRS in the region of the tumor is presumably both in the cellular space and in the interstitial fluid space. To gain a better perspective on the possible distribution of 5-FU in these two spaces, we performed a simulation/deconvolution of the experimental data of patient no. 42.

View larger version (17K): [in this window] [in a new window]   Fig 4. A five-compartment conceptual model of the biodistribution and metabolism of 5-FU in a tumor after systemic administration of this drug.

View larger version (33K): [in this window] [in a new window]   Fig 5. Deconvolution of the 5-FU signal obtained noninvasively from St. Vincent patient no. 42 after administration of 5-FU alone, performed using ADAPT-II and the model of Fig 4. The experimental data (total 5-FU seen by coil) were fitted to a two-exponential equation.

Although both IFN-2a and MTX were found in this study to increase the t_1/2 of 5-FU in the tumor, there may be more than one mechanistic explanation for it. We postulate that MTX not only stimulates the formation of nucleotides by increasing the availability of ribosyl-pyrophosphate, but it also has an effect on the transport of 5-FU across the cell membrane. This effect was enhanced by three orders of magnitude in our animal studies.¹⁸ It will, therefore, be interesting to determine whether IFN-2a and MTX enhance the membrane transport of 5-FU or inhibit its efflux or whether its effect may be dependent on action at the interstitial fluid compartment level. In the case of MTX, we have already shown that the drug has a significant effect on increasing the t_1/2 of free 5-FU, and that the lumped transfer constant of free 5-FU from the tumoral space had decreased by three orders of magnitude.¹⁸

In conclusion, these results pharmacologically document the in vivo chemotherapeutic modulation of 5-FU biodistribution by two distinct modulators, IFN-2a and MTX. The present study has also demonstrated that ¹⁹F-MRS is a useful method for noninvasively monitoring, in individual patients, the possible chemotherapeutic effectiveness of 5-FU. It also shows that the efficacy of a putative 5-FU modulator can be studied by ¹⁹F-MRS pharmacokinetic analysis and that such information may assist in the development of rational therapeutic strategies. Moreover, the use of pharmacokinetic methods and the analysis of noninvasive data, obtained using MRS, may provide unique and novel information that might be able to guide the oncologist toward more efficient planning of tumor chemotherapy.

	ACKNOWLEDGMENTS

 
The 5-FU + IFN part of this study was supported, in part, by a gift from Schering Inc, Kenilworth, NJ; the SWOG 9118 study was supported by grants no. CA 58928, CA 35200, and CA 56138 from the United States Public Health Service (Bethesda, MD).

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Pinedo HM, Peters GJ: Fluorouracil: Biochemistry and pharmacology. J Clin Oncol 6:1653-1664, 1988[Abstract/Free Full Text]

2. Wolf W, Presant CA, Servis KL, et al: Tumor trapping of 5-FU: In vivo 19F-NMRS pharmacokinetics in tumor-bearing humans and rabbits. Natl Acad Sci USA 87:492-496, 1990[Abstract/Free Full Text]

3. Presant CA, Wolf W, Albright MJ, et al: Human tumor 5-FU trapping: Clinical correlations of in vivo 19F -NMRS pharmacokinetics. J Clin Oncol 8:1868-1873, 1990[Abstract]

4. Presant CA, Wolf W, Waluch V, et al: Association of intratumoral pharmacokinetics of 5-FU with clinical response. Lancet 343:1184-1187, 1994[Medline]

5. Rustum YM, Cao S, Zhang Z: Rationale for treatment design: Biochemical modulation of 5-fluorouracil by leucovorin. Cancer J Sci Am 4:12-18, 1998[Medline]

6. Labianca R, Pessi A, Facendola G, et al: Modulated 5-fluorouracil (5-FU) regimens in advanced colorectal cancer: A critical review of comparative studies. Eur J Cancer 32A:S7-12, 1996 (suppl 5)

7. Ardalan B, Luis R, Jaime M, et al: Biomodulation of fluorouracil in colorectal cancer. Cancer Invest 16:237-251, 1998[Medline]

8. Horowitz R, Schwartz EL, Wadler S: Modulation of 5-fluorouracil by interferon: A review of potential cellular targets. Med Oncol 12:3-8, 1995[Medline]

9. Houghton JA, Morton CL, Adkins DA, et al: Locus of interaction among 5-FU leucovorin and interferon-2a in colon carcinoma cells. Cancer Res 53:4243-4250, 1993[Abstract/Free Full Text]

10. Lee M, Price D, Specht S, et al: Interferon modulation of 5-FU: Use in neoadjuvant therapy inhibits experimental liver metastases in nude mice. Drugs 3:413-418, 1992

11. Fornasiero A, Daniele O, Giotto C, et al: Interferon-2 and 5-FU in advanced colorectal cancer. Tumori 76:385-388, 1990[Medline]

12. Schwartz LE, Baptiste N, O’Connor CJ, et al: Potentiation of the anti-tumor activity of 5-FU in colon carcinoma cells by the combination of interferon and deoxyribonucleosides results from complementary effects on thymidine phosphorylation. Cancer Res 54:1472-1478, 1994[Abstract/Free Full Text]

13. Kocha W: 5-FU plus interferon-2a versus 5-FU plus leucovorin in metastatic colorectal cancer: Results of a multicenter multinational phase III study. Proc Am Soc Clin Oncol 12:193, 1993 (abstr 562)

14. York M, Greco FA, Figlin RA, et al: A randomized phase III trial comparing 5-FU with or without interferon-2a for advanced colorectal cancer. Proc Am Soc Clin Oncol 12:200, 1993 (abstr 590)

15. Machov D: The pharmacological modulation of 5-FU with folinic acid methotrexate trimetrexate and m-phosphonacetyl-1-aspartic acid (PALA) mechanisms of the interactions and clinical data. Bull Cancer 9:74s-78s, 1994 (suppl 2)

16. Cadman E, Heimer R, Davis L: Enhanced 5-FU nucleotide formation after MTX administration: Explanation for drug synergism. Science 205:1135-1137, 1979[Abstract/Free Full Text]

17. Marsh JC, Bertino JR, Katz KH, et al: The influence of drug interval on the effect of MTX and 5-FU in the treatment of advanced colorectal cancer. J Clin Oncol 9:371-380, 1991[Abstract]

18. El-Tahtawy A, Wolf W: In vivo measurements of intratumoral metabolism modulation and pharmacokinetics of 5-FU using ¹⁹F-NMRS. Cancer Res 51:5806-5812, 1991[Abstract/Free Full Text]

19. Katzir I, Berman E, Wolf W, et al: Enhancement of 5-FU anabolism by MTX and TMTX in two rat solid tumor models: Walker 256 carcinosarcoma in vitro and Novikoff hepatoma in vivo evaluated by ¹⁹F-MRS. Invest 18:20-27, 2000

20. Rosenfeld D, Zur Y: Design of adiabatic selective pulses using optimal control theory. Magn Reson Med 36:401-409, 1996[Medline]

21. Green S, Weiss GR: Southwest Oncology Group standard response criteria endpoint definitions and toxicity criteria. Invest New Drugs 10:239-253, 1992[Medline]

Submitted January 4, 1999; accepted August 17, 1999.

This article has been cited by other articles:

				N. Panayotatos Drug-binding Cavities in Long-Lived Biologics: Cause for Concern but Also Potential Benefit J. Clin. Pharmacol., October 1, 2008; 48(10): 1208 - 1211. [Abstract] [Full Text] [PDF]

				J. R. Bading, P. B. Yoo, J. D. Fissekis, M. M. Alauddin, D. Z. D'Argenio, and P. S. Conti Kinetic Modeling of 5-Fluorouracil Anabolism in Colorectal Adenocarcinoma: A Positron Emission Tomography Study in Rats Cancer Res., July 1, 2003; 63(13): 3667 - 3674. [Abstract] [Full Text] [PDF]

				J. Pehoushek and K. J. Smith Imiquimod and 5% Fluorouracil Therapy for Anal and Perianal Squamous Cell Carcinoma In Situ in an HIV-1-Positive Man Arch Dermatol, January 1, 2001; 137(1): 14 - 16. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Presant, C. A. Articles by Shani, J. Search for Related Content PubMed PubMed Citation Articles by Presant, C. A. Articles by Shani, J.