Impact of Severe Hypothyroidism on

CASE REPORTS Impact of Severe Hypothyroidism on Cyclophosphamide Disposition and Routes of Metabolism and Transport in a Patient with Treatment-Resistant Lupus Nephritis So Yoon Jang, Mary Anne Dooley, Melanie S Joy

C

yclophosphamide belongs to the oxazophorine class of alkylating agents with antitumor and immunomodulating activities and is used as standard induction treatment of lupus nephritis.1-3 Patient responses to therapy are highly variable, with outcomes ranging from complete remission to treatment resistance.4-7 Several mechanisms for treatment resistance to cyclophosphamide include alterations in its metabolism and cellular responses to DNA injury. Cyclophosphamide is a prodrug that requires metabolic activation by CYP3A4, CYP2C9, and CYP2B6 to the 4-hydroxycyclophosphamide metabolite.8-10 Several inactive and toxic metabolites are also formed.11 The exposure to cyclophosphamide and 4-hydroxycyclophosphamide can be influenced by mediators that influence the expression and activity of the relevant CYP Author information provided at end of text.

© 1967-2013 Harvey Whitney Books Co. All rights reserved. No part of this document may be reproduced or transmitted in any form or by any means without prior written permission of Harvey Whitney Books Co. For reprints of any article appearing in The Annals, please contact [email protected]

theannals.com

OBJECTIVE: To report what we believe to be the first case of severe hypothyroidism with reduced drug metabolism and transport activity.

CASE SUMMARY: A 32-year-old African American woman with a history of treatmentresistant lupus nephritis and concurrent hypothyroidism was participating in a clinical study to evaluate cyclophosphamide pharmacokinetics in patients with glomerulonephritis due to lupus nephritis and small-vessel vasculitis. Thyroid-stimulating hormone levels ranged from 60 to 300 µIU/mL, despite high doses of thyroid replacement hormone (levothyroxine 400 µg twice weekly). The pharmacokinetics of the probe drug cocktail (flurbiprofen/fexofenadine) were altered, with formation clearance of flurbiprofen (CYP2C9 function) lower in our patient versus the average value in our study cohort, suggesting a reduction in activity. The area under the concentration-time curve from 0 to 24 hours for fexofenadine (transporter function) was 2-fold higher in our patient compared to that of other study patients. Pharmacokinetic data showed markedly decreased cyclophosphamide clearance and exposure to 4-hydroxycyclophosphamide, as well as a reduced metabolic ratio of 4hydroxycyclophosphamide to cyclophosphamide.

DISCUSSION: Previous cases of altered pharmacokinetics and toxicity of medications in patients with mild to moderate thyroid dysfunction have been published. Our case evaluated the impact of a severe form of hypothyroidism on cyclophosphamide pharmacokinetics and probe drug metabolism and transport. If changes were not demonstrated at the extreme spectrum of hypothyroidism, there would be little concern for changes in patients with less severe disease. Profound hypothyroidism likely contributed to the patient’s poor response to cyclophosphamide treatment through its influence on CYP isoenzymes responsible for the activation to 4hydroxycyclophosphamide and possibly through reduced transport function.

CONCLUSIONS: Clinicians should monitor for significant hypothyroidism in patients who are prescribed drugs (eg, cyclophosphamide) that require metabolic conversion to form active therapeutic moieties.

Ann Pharmacother 2013;47:e35.

Published Online, 4 Jun 2013, theannals.com, doi: 10.1345/aph.1S012

The Annals of Pharmacotherapy

Downloaded from aop.sagepub.com by guest on October 11, 2013

■

2013 July/August, Volume 47

■

e35

SY Jang et al.

enzymes. It is unknown which systemic transporters influence the disposition of cyclophosphamide and its metabolite. Chronic kidney disease has been documented to alter the disposition of cyclophosphamide and 4-hydroxycyclophosphamide.12 Altered pharmacokinetics of additional drugs has also been reported by our group and others in patients with glomerulonephritis secondary to lupus nephritis and small-vessel vasculitis.13-15 Extensive reviews of the metabolism and pharmacokinetics of cyclophosphamide have been published.11,16 Animal studies have suggested that thyroid hormone status influences the activity of cytochrome enzymes, nicotinamide adenine dinucleotide phosphate (NADPH)cytochrome P450 reductase, and P-glycoprotein.17-20 There are published cases of altered pharmacokinetics and toxicity of various medications including cyclosporine, tacrolimus, and oxazepam in patients with thyroid dysfunction.21-24 We report the impact of a severe form of hypothyroidism on cyclophosphamide pharmacokinetics, as well as on the metabolism and transport of probe drugs. The rationale behind publishing this case was if changes were not demonstrated at the extreme spectrum of hypothyroidism, there would be little concern for patients with less severe disease. Furthermore, the patient described here was part of a study; thus, factors influencing the role of lupus nephritis in altered pharmacokinetics through components of metabolism and transport could be controlled for, limiting the results to the influence of severe hypothyroidism. To our knowledge, there have been no published reports describing alterations in cyclophosphamide pharmacokinetics in association with thyroid dysfunction. Case Report

A 32-year-old African American female with a history of treatment-resistant lupus nephritis and concurrent hypothyroidism was participating in a clinical study to evaluate cyclophosphamide pharmacokinetics in patients with glomerulonephritis due to lupus nephritis and small-vessel vasculitis.13 As part of the study, to assess the activity of targeted routes of drug metabolism and transport, the patient received a combination one-time cocktail of drugs at the time of the cyclophosphamide administration. This cocktail consisted of fexofenadine 60 mg to assess transporter function, flurbiprofen 50 mg to assess CYP2C9 activity, and intravenous 14C-erythromycin 3 µCi to assess the activity of CYP3A4. The patient was enrolled and admitted for a 72-hour inpatient stay under the study protocol. The patient was diagnosed with lupus nephritis class V in 2006 by renal biopsy. At that time she was found to have nephrotic-range proteinuria (4.7 g/day) with normal serum creatinine level (0.9 mg/dL). Her treatment history for lupus nephritis included oral mycophenolate mofetil, e35

■


■

intravenous cyclophosphamide (2006), and intravenous rituximab (2007). Previous treatments had failed to induce complete remission. The patient had continued to have persistent nephrotic-range proteinuria. Treatment with oral mycophenolate mofetil was discontinued 4 months prior to the study, at the initiation of the current course of cyclophosphamide. Hypothyroidism was diagnosed in 2006 when the patient presented with profound weakness. Thyroid-stimulating hormone (TSH) levels had ranged from 60 to 300 µIU/mL (reference range 0.6-30) since the diagnosis, despite the patient receiving high oral doses of thyroid replacement hormones. The TSH level was 100-fold higher than the targeted normal reference range. A clinician previously treating the patient suspected poor oral absorption of levothyroxine because of the persistently elevated TSH and prescribed intermittent intramuscular injections of levothyroxine up to 400 µg twice weekly. At the time of the research study, the patient was receiving intramuscular levothyroxine 400 µg twice weekly and reported symptoms of extreme lethargy and dysarthria. Additional pertinent medical history included hypertension (blood pressure 142/101 mm Hg), obesity (weight 122 kg), osteoarthritis, and obstructive sleep apnea treated with continuous positive airway pressure. On the first day of the cyclophosphamide pharmacokinetics research study, the patient received intravenous cyclophosphamide 1.1 g (0.5 g/m2) over 1 hour and the probe drug cocktail (intravenous 14C-erythromycin 3 µCi, flurbiprofen 50 mg, and oral fexofenadine 60 mg). As part of the study, blood samples were drawn prior to probe drug cocktail administration and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 18, 24, 36, 48, and 72 hours. Urine was collected during the following intervals: 0-6, 6-12, 12-24, 24-36, 36- 48, and 48-72 hours. Treatment prior to cyclophosphamide administration included oral ondansetron 24 mg, intravenous prehydration with sodium chloride 100 mL/h, and intravenous mesna 0.2 g prior to and following cyclophosphamide at 3-hour intervals, for 4 doses. These treatments were consistent between our patient and the other study patients. During the study, our patient continued to receive her prescribed daily oral medications of aspirin 81 mg, amlodipine 10 mg, valsartan 320 mg, enalapril 10 mg, simvastatin 40 mg, and azelastine nasal spray. Medications prescribed on an asneeded basis included hydrocodone/acetaminophen, oxycodone/acetaminophen, and diphenhydramine. All study patients were maintained on a CYP450 diet (exclusion of cruciferous vegetables, spinach, garlic, grapefruit, chargrilled meats, and smoked meats) throughout the study.25 The patient’s clinical data on the day of the study included serum creatinine 0.9 mg/dL, creatinine clearance (CrCl) 140 mL/min, urinary protein-creatinine ratio 4, and serum albumin 3.3 g/L. TSH was 339 µIU/mL (reference range 0.6-3.30) and unbound thyroxine (T4) was 0.28 ng/dL (0.71-1.40). On the day of the study but several hours after


theannals.com

Impact of Hypothyroidism on Cyclophosphamide in Treatment-Resistant Lupus Nephritis

the administration of cyclophosphamide and phenotyping drugs, the patient received levothyroxine 400 µg by the intramuscular route. A 2-day follow-up TSH level of 181 µIU/mL and unbound T4 level of 0.58 ng/dL were reported by the hospital laboratory. Confirmation of the low unbound T4 level by direct equilibrium dialysis resulted in a level of 0.4 ng/dL (reference range 0.8-2.0). A thyroidbinding globulin concentration was reported to be 25 µg/mL (11-27) and a morning cortisol level was 19.5 µg/dL (4.5-22.7). The levothyroxine therapy was modified to 400 µg daily by the oral route. The patient was examined 1 year after her study participation. Follow-up review showed that she had failed to achieve remission after this course of intravenous cyclophosphamide and continued to demonstrate persistent proteinuria of 3 g/day, with an unchanged creatinine level of 0.86 mg/dL. Anti– double-strand DNA was continually detected at 1:10 and serum complements were low, with C3 80 mg/dL (reference range 88-171) and C4 less than 8 mg/dL (15-48). Pharmacokinetics

Noncompartmental pharmacokinetics were analyzed for the fexofenadine and flurbiprofen cocktail drugs and for cyclophosphamide and 4-hydroxycyclophosphamide. The activity of the CYP3A4 metabolism pathway was assessed by the erythromycin breath test. The following pharmacokinetics data were reported for the case patient and mean (SD) values were reported for the patients with lupus nephritis who had normal thyroid function: maximum plasma concentration after a dose (Cmax), time to maximum plasma concentration, half-life, area under the plasma concentration-time curve from 0 to infinity (AUC0-∞) and 0 to 24 hours (AUC0-24), and apparent oral clearance (Cl/F). For flurbiprofen (CYP2C9 probe), formation clearance (Clf) was also reported. The metabolic ratio of metabolite AUCparent drug AUC was computed for flurbiprofen and cyclophosphamide. For fexofenadine (transporter probe), AUC was reported. The erythromycin probe was used to assess CYP3A4 activity and is reported as the percentage of the dose that was metabolized per hour. The Clf of flurbiprofen (4-hydroxyflurbiprofen/flurbiprofen in the urine) was slightly lower in our patient versus the mean value in the study cohort (0.48 vs 0.58 [0.23] L/h). The Cl/F of flurbiprofen was similar between our patient and the study cohort. The AUC0-24 for fexofenadine was 2-fold higher in our patient compared to the other study patients (1908 vs 1070 [553] µg•h/L). CYP3A4 activity, as assessed by erythromycin, was similar between our patient and the study cohort. Cyclophosphamide Cl/F was 2-fold lower (4.7 vs 11.6 [5.1] L/h) and the 4-hydroxycyclophosphamide AUC0-∞ was 2-fold lower (2707 vs 4490 [2484] ng•h/mL) in our patient versus the remainder of the study cohort. theannals.com

Discussion

Clinicians have been trained to be aware of the possibility of altered drug responses in patients with thyroid disorders. In these patients, studies have reported altered responses and adverse events to drugs including digoxin, oxazepam, and warfarin.26 Our case prompted the question of how profoundly clinical hypothyroidism can influence the pharmacokinetics of cyclophosphamide and individual pathways of drug metabolism and transport. It assessed the pharmacokinetics of cyclophosphamide, a drug that requires metabolic activation through CYP enzymes to the active 4-hydroxycyclophosphamide, in patients with severe hypothyroidism, compared to that in a similar group of patients who were in the euthyroid state. Since metabolism of cyclophosphamide is required for exposure to the active metabolite, 4-hydroxycyclophosphamide, differing treatment responses could be attributed to the success of this conversion process. Because our patient had extremely elevated TSH levels, the activity of drug metabolism and transport pathways in this case should represent the most severe end of the spectrum, above and beyond the changes in activity that would be realized in patients with lupus nephritis. Our patient was receiving intravenous cyclophosphamide, so while gastrointestinal transporters may not be relevant, liver and kidney transport function may be important for systemic exposure to cyclophosphamide and 4-hydroxycyclophosphamide. Because several CYP pathways (eg, CYP2B6, CYP3A4, and CYP2C9) are important in the activation of cyclophosphamide to 4-hydroxycyclophosphamide, the relative activity and importance of each route could be assessed in severe hypothyroidism.8-10 CYP3A4 and CYP2C9 contribute 25% and 10%, respectively, to cyclophosphamide’s metabolism, amounting to 35% of the activity being evaluated. Although CYP2B6 contributes to 40% of cyclophosphamide’s total metabolism, the activity of this pathway was not assessed in our patient because of the loss of the plasma sample.16,27,28 Our patient demonstrated markedly reduced mean cyclophosphamide clearance (4.7 vs 11.6 [5.1] L/h) and a resultant decrease in mean exposure (AUC0-∞) to 4-hydroxycyclophosphamide (2707 vs 4490 [2484] ng•h/mL). The mean metabolic ratio (4-hydroxycyclophosphamide-cyclophosphamide AUC) was reduced 2-fold in our patient as compared to the lupus nephritis study cohort (0.02 vs 0.05 [0.03]). Clinically, our patient failed to have a response to intravenous cyclophosphamide during the treatment course. Our patient also failed to have a clinical response after a prior course of cyclophosphamide. The activity of the CYP2C9 pathway was reduced in our patient versus the study cohort, as reflected in the Clf of flurbiprofen (0.48 vs 0.58 [0.23] L/h). This could have implications for other drugs that require the CYP2C9 pathway for metabolism, including cyclophos-


■


■

e35

SY Jang et al.

phamide. The reduction in metabolism of flurbiprofen and cyclophosphamide, and transport of fexofenadine, are convincingly attributable to the profound hypothyroid state as opposed to kidney dysfunction, given that our patient had well-preserved kidney function as assessed by CrCl (140 mL/min). The other lupus nephritis patients in the study had a slightly lower level of kidney function (CrCl 95.1 [38.1] mL/min) and yet had greater CYP2C9 and transporter function, ruling out kidney function as a contributing variable in our patient. Since our patient was not receiving other medications that were different from those of the study cohort, this cannot be a factor in explaining altered metabolism or transport function. We can hypothesize that a reduction in CYP2B6 activity could have also contributed to reduced activation to 4-hydroxycyclophosphamide. From a pathophysiology perspective, many of the CYP450 isoenzymes are under the control of hormones, including the thyroid hormones. Thyroid hormones contribute to the regulation of CYP450 activity through the expression and function of NADPH P450 reductase in the liver and extrahepatic tissues.18 A study demonstrated an 85% reduction in liver NADPH P450 reductase activity in rats that were depleted of circulating T4 by hypophysectomy or methimazole treatment.18 This loss of liver CYP450 reductase activity was fully reversed by T4 replacement. The study also demonstrated induction of CYP450 reductase activity in the kidneys, adrenal gland, and heart. Transporter function, as assessed by fexofenadine, was reduced in our patient as reflected by an AUC0-24 that was 2-fold higher compared to that of other study patients (1908 vs 1070 [553] µg•h/L). Reduced transporter function would lead to increased plasma concentrations of other therapeutic agents that are substrates of the P-glycoprotein and organic anion transporting polypeptide transporters. Studies have reported increased P-glycoprotein expression after exposure to thyroid hormones,29,30 suggesting that a deficient state would have decreased expression of P-glycoprotein.29,30 In vitro studies have reported upregulation of intestinal P-glycoprotein expression in Caco-2 cells under the influence of T3 and T4.31 Our case demonstrates altered activity of CYP450 and transporter pathways in a patient with profound hypothyroidism. The reduction in metabolic conversion of cyclophosphamide to 4-hydroxycyclophosphamide likely contributed to the lack of therapeutic response to cyclophosphamide therapy. To our knowledge, this is the first published case to demonstrate this clinical finding. Clinicians should monitor for hypothyroidism in patients who are prescribed drugs such as cyclophosphamide that rely on CYPs for activation and deactivation. So Yoon Jang MD, Nephrology Fellow, School of Medicine, Divi-

sion of Nephrology and Hypertension, University of North Carolina, UNC Kidney Center, Chapel Hill

e35

■


■

Mary Anne Dooley MD, Associate Professor, School of Medicine, Division of Nephrology and Hypertension, University of North Carolina, UNC Kidney Center Melanie S Joy PharmD PhD FCCP FASN, Assistant Professor, Skaggs School of Pharmacy and Pharmaceutical Sciences, Aurora, CO Correspondence: Dr. Joy, [email protected] Reprints/Online Access: www.theannals.com/cgi/reprint/aph.1S012 Conflict of interest: Authors reported none

Funding: This research was funded by the National Institutes of Health K23DK64888 (MSJ), General Clinical Research Centers program of the Division of Research Resources, National Institutes of Health RR00046 (MSJ), Clinical and Translational Science Award U54RR024383 (MSJ), and American College of Clinical Pharmacy Research Institute’s Frontier’s Award (MSJ) © 1967-2013 Harvey Whitney Books Co. All rights reserved. No part of this document may be reproduced or transmitted in any form or by any means without prior written permission of Harvey Whitney Books Co. For reprints of any article appearing in The Annals, please contact [email protected]

References

1. Austin HA III, Klippel JH, Balow JE, et al. Therapy of lupus nephritis. Controlled trial of prednisone and cytotoxic drugs. N Engl J Med 1986;314: 614-9. 2. Boumpas DT, Austin HA III, Vaughn EM, et al. Controlled trial of pulse methylprednisolone versus two regimens of pulse cyclophosphamide in severe lupus nephritis. Lancet 1992;340:741-5. 3. Gourley MF, Austin HA III, Scott D, et al. Methylprednisolone and cyclophosphamide, alone or in combination, in patients with lupus nephritis. A randomized, controlled trial. Ann Intern Med 1996;125:549-57. 4. Bakir AA, Levy PS, Dunea G. The prognosis of lupus nephritis in AfricanAmericans: a retrospective analysis. Am J Kidney Dis 1994;24:159-71. 5. Dooley MA, Hogan S, Jennette C, Falk R. Cyclophosphamide therapy for lupus nephritis: poor renal survival in black Americans. Glomerular Disease Collaborative Network. Kidney Int 1997;51:1188-95. 6. Ludeman SM, Gamcsik MP. Mechanisms of resistance against cyclophosphamide and ifosfamide: can they be overcome without sacrificing selectivity? Cancer Treat Res 2002;112:177-97. 7. Vyas S, Hidalgo G, Baqi N, Von Gizyki H, Singh A. Outcome in African-American children of neuropsychiatric lupus and lupus nephritis. Pediatr Nephrol 2002;17:45-9. 8. Chang TK, Weber GF, Crespi CL, Waxman DJ. Differential activation of cyclophosphamide and ifosphamide by cytochromes P- 450 2B and 3A in human liver microsomes. Cancer Res 1993;53:5629-37. 9. Huang Z, Roy P, Waxman DJ. Role of human liver microsomal CYP3A4 and CYP2B6 in catalyzing N-dechloroethylation of cyclophosphamide and ifosfamide. Biochem Pharmacol 2000;59:961-72. 10. Roy P, Tretyakov O, Wright J, Waxman DJ. Stereoselective metabolism of ifosfamide by human P- 450s 3A4 and 2B6. Favorable metabolic properties of R-enantiomer. Drug Metab Dispos 1999;27:1309-18. 11. Zhang J, Tian Q, Yung Chan S, et al. Metabolism and transport of oxazaphosphorines and the clinical implications. Drug Metab Rev 2005;37:611703. 12. Haubitz M, Bohnenstengel F, Brunkhorst R, Schwab M, Hofmann U, Busse D. Cyclophosphamide pharmacokinetics and dose requirements in patients with renal insufficiency. Kidney Int 2002;61:1495-501. 13. Joy MS, La M, Wang J, et al. Cyclophosphamide and 4-hydroxycyclophosphamide pharmacokinetics in patients with glomerulonephritis secondary to lupus and small vessel vasculitis. Br J Clin Pharmacol 2012;74:445-55. 14. de Miranda Silva C, Fernandes BJ, Donadi EA, et al. Influence of glomerular filtration rate on the pharmacokinetics of cyclophosphamide enantiomers in patients with lupus nephritis. J Clin Pharmacol 2009;49:965-72. 15. Belfayol-Pisante L, Guillevin L, Tod M, Fauvelle F. Pharmacokinetics of cyclophosphamide (CP) and 4-OH-CP/aldophosphamide in systemic vasculitis. Fundam Clin Pharmacol 2000;14:415-21.


theannals.com

Impact of Hypothyroidism on Cyclophosphamide in Treatment-Resistant Lupus Nephritis

16. Zhang J, Tian Q, Chan SY, Duan W, Zhou S. Insights into oxazaphosphorine resistance and possible approaches to its circumvention. Drug Resist Updat 2005;8:271-97. 17. Burk O, Brenner SS, Hofmann U, et al. The impact of thyroid disease on the regulation, expression, and function of ABCB1 (MDR1/P glycoprotein) and consequences for the disposition of digoxin. Clin Pharmacol Ther 2010; 88:685-94. 18. Ram PA, Waxman DJ. Hepatic P450 expression in hypothyroid rats: differential responsiveness of male-specific P450 forms 2a (IIIA2), 2c (IIC11), and RLM2 (IIA2) to thyroid hormone. Mol Endocrinol 1991;5:13-20. 19. Webb SJ, Xiao GH, Geoghegan TE, Prough RA. Regulation of CYP4A expression in rat by dehydroepiandrosterone and thyroid hormone. Mol Pharmacol 1996;49:276-87. 20. Lanni A, Moreno M, Lombardi A, Goglia F. Rapid stimulation in vitro of rat liver cytochrome oxidase activity by 3,5-diiodo-L-thyronine and by 3,3¢-diiodo-L-thyronine. Mol Cell Endocrinol 1994;99:89-94. 21. Leong SO, Lye WC, Tan CC, Lee EJ. Acute cyclosporine A nephrotoxicity in a renal allograft recipient with hypothyroidism. Am J Kidney Dis 1995;25:503-5. 22. Nishimura M, Yamada K, Matsushita K, et al. Changes in trough levels of whole blood cyclosporine and graft function of a kidney transplant recipient with onset of hypothyroidism after transplantation. Transplantation 1996;62:1509-11.

theannals.com

23. Haas M, Kletzmayer J, Staudinger T, et al. Hypothyroidism as a cause of tacrolimus intoxication and acute renal failure: a case report. Wien Klin Wochenschr 2000;112:939-41. 24. Scott AK, Khir AS, Bewsher PD, Hawksworth GM. Oxazepam pharmacokinetics in thyroid disease. Br J Clin Pharmacol 1984;17:49-53. 25. Fujita K. Food-drug interactions via human cytochrome P450 3A (CYP3A). Drug Metabol Drug Interact 2004;20:195-217. 26. O’Connor P, Feely J. Clinical pharmacokinetics and endocrine disorders. Therapeutic implications. Clin Pharmacokinet 1987;13:345-64. 27. Chen CS, Lin JT, Goss KA, He YA, Halpert JR, Waxman DJ. Activation of the anticancer prodrugs cyclophosphamide and ifosfamide: identification of cytochrome P450 2B enzymes and site-specific mutants with improved enzyme kinetics. Mol Pharmacol 2004;65:1278-85. 28. Xie HJ, Yasar U, Lundgren S, et al. Role of polymorphic human CYP2B6 in cyclophosphamide bioactivation. Pharmacogenomics J 2003;3:53-61. 29. Nishio N, Katsura T, Inui K. Thyroid hormone regulates the expression and function of P-glycoprotein in Caco-2 cells. Pharm Res 2008;25:1037-42. 30. Siegmund W, Altmannsberger S, Paneitz A, et al. Effect of levothyroxine administration on intestinal P-glycoprotein expression: consequences for drug disposition. Clin Pharmacol Ther 2002;72:256-64. 31. Mitin T, Von Moltke LL, Court MH, Greenblatt DJ. Levothyroxine upregulates P-glycoprotein independent of the pregnane X receptor. Drug Metab Dispos 2004;32:779-82.


■


■

e35