Share this article on:

Thalidomide Use and Outcomes in Pediatric Patients With Crohn Disease Refractory to Infliximab and Adalimumab

Felipez, Lina M.; Gokhale, Ranjana; Tierney, Matthew P.; Kirschner, Barbara S.

Journal of Pediatric Gastroenterology and Nutrition: January 2012 - Volume 54 - Issue 1 - p 28–33
doi: 10.1097/MPG.0b013e318228349e
Original Articles: Gastroenterology

Objective: The aim of the study was to evaluate thalidomide as rescue therapy for pediatric patients with severe refractory Crohn disease (CD) who failed to respond to antitumor necrosis factor (TNF) biologic agents.

Patients and Methods: A computerized database was used to identify children with CD who had failed conventional immunosuppression therapy and received thalidomide rescue therapy. Twelve patients, mean age at diagnosis 10 years, were identified. Eight children had disease localized to the ileum and colon and 4 to the gastroduodenal area and colon. Five cases were complicated by strictures and 7 by fistulae. Previous drug therapy included azathioprine/6-mercaptopurine (11/12), methotrexate (7/12), and anti-TNF biologics (12/12). Outcome measures were Harvey-Bradshaw Index, change in prednisone dose, hospitalizations, bowel resections, and incision and drainage procedures. Laboratory evaluations were calculated before and after 1 to 6 months of thalidomide.

Results: Mean Harvey-Bradshaw Index score improved from 11.8 to 3.9 (P = 0.0004), mean prednisone dose decreased from 13.9 to 2.3 mg/day (P = 0.001), mean number of hospitalizations decreased from 6.3 to 1.3 (P = 0.002), and erythrocyte sedimentation rate decreased from 35 to 14 mm/h (P = 0.02). The surgery rate pre-thalidomide was 0.031 and on thalidomide was 0.004. Of the 7 patients with fistulae, 5 had complete fistula closure, 1 had partial closure, and 1 showed no improvement. Adverse reactions that resulted in discontinuation of thalidomide are as follows: 42% peripheral neuropathy, 17% worsening of the CD, 8% dizziness, and 8% allergic reaction. All 5 patients who developed peripheral neuropathy had clinical resolution of the neurologic symptoms within 2 to 3 months after stopping thalidomide.

Conclusions: Thalidomide is a potentially effective rescue therapy for severe refractory CD in children who fail to respond to anti-TNF medications.

Section of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, University of Chicago Comer Children's Hospital, Chicago, IL.

Address correspondence and reprint requests to Lina M. Felipez, MD, University of Chicago Medical Center, 5841 S Maryland Ave, MC-4065, Chicago, IL 60637 (e-mail:

Received 18 January, 2011

Accepted 3 June, 2011

The authors report no conflicts of interest.

Thalidomide (α-phthalimidoglutarimide) is a derivative of glutamic acid, which is formulated as a 1:1 racemic mixture of S and R isomers. The S isomer provides immunomodulatory effects, whereas the R isomer has the sedative effects and is associated with teratogenicity (phocomelia and amelia) (1). Thalidomide is well absorbed orally, maximal plasma levels being reached in 2 to 4 hours. The half-life is 5 hours, with minimal excretion via hepatic or renal pathways.

Thalidomide was introduced in Germany in 1956 and in the rest of Europe, Australia, Canada, and South America in 1957. It was widely used by pregnant women for the treatment of morning sickness. In 1961, the drug was withdrawn, leaving between 6000 and 10,000 affected children with birth defects characterized by phocomelia or amelia (2).

In 1965, Sheshkin et al (3) administered thalidomide for its sedative effect to patients with leprosy and observed virtually complete resolution of patients’ lepromatous cutaneous manifestations. In 1991, Sampaio et al (4) demonstrated the antitumor necrosis factor (TNF)-α activity of thalidomide in in vitro studies. In 1998, the Food and Drug Administration approved thalidomide to treat the cutaneous manifestations of leprosy (erythema nodosum leprosum) (5).

Multiple studies have shown that diseases in which TNF-α is a mediator of inflammation respond to thalidomide treatment, including HIV-associated oral and esophageal aphthous ulcerations and cachexia, Behçet disease, cutaneous lupus erythematosus, pyoderma gangrenosum, rheumatoid arthritis, ankylosing spondylitis, sarcoidosis, systemic lupus erythematosis, Sjogren syndrome, multiple myeloma, graft-versus-host disease, and some solid tumors (3,6–10).

Thalidomide has 3 postulated mechanisms of action: immunomodulatory effects, inhibition of angiogenesis, and interference with cell adhesion. Immunologic mechanisms include inhibition of the synthesis of TNF-α, interference with cell adhesion by promoting the degradation of TNF messenger RNA, decreased TNF-α release from activated monocytes, reduced nuclear factor-κB activation by suppressing I-κB kinase activity, switch in cytokine production from TH1 to TH2 profile, inhibition of interleukin (IL)-12 synthesis (a potent proinflammatory cytokine), decreased interferon-γ synthesis, enhanced IL-4 production (essential for development of TH2 pathway), and IL-10 production (a proinflammatory cytokine). Inhibition of angiogenesis results from reduced synthesis of vascular endothelial-derived growth factor and basic fibroblast growth factor. Interference with cell adhesion is caused by reduced expression of intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 (11–13).

There are few published studies that address the role of thalidomide in treating Crohn disease (CD) in adults and children (14–19). This is the first study, to our knowledge, that evaluated thalidomide in children and adolescents with severe refractory CD, all of whom had failed anti-TNF biologics. This is also the first analysis of the effect of thalidomide in pediatric patients with fistulizing CD.

Back to Top | Article Outline


A computerized database of patients with pediatric-onset inflammatory bowel disease at the University of Chicago was used to identify all of the children and adolescents with severe refractory steroid-dependent CD who failed conventional medications including immunosuppression therapy (5-aminosalicylic acid, 6-mercaptopurine/azathioprine, methotrexate, anti-TNF biologics) and subsequently received thalidomide rescue therapy (Table 1). The study was approved by the Institutional Review Board of the University of Chicago.



Data were obtained retrospectively from the patients’ medical records from January 2001 to January 2010 and reviewed independently by 1 investigator (L.M.F.). Data collected included demographic information (eg, age, sex, ethnicity); medical history; CD history including disease location and behavior; radiologic studies including rectal and pelvic MRI, small bowel follow-through x-ray (SBFT), computed tomography scans, and magnetic resonance enterography; previous medications and surgical procedures (incision and drainage, setons, bowel resections); and hospitalizations. Patients were evaluated in the Pediatric Inflammatory Bowel Disease Clinics at the University of Chicago every 4 to 6 weeks. The patients’ clinical symptoms, medications, physical examination, and weight and height were recorded in each visit. A comprehensive neurologic history was taken during each visit. For fistulizing disease, fistula number and drainage were assessed by the patient's gastroenterologist. Routine laboratory tests were ordered as per the attending physician.

Adverse events that occurred during thalidomide therapy were recorded at each clinic visit. Patients were specifically questioned about neurologic symptoms (paresthesias, pain, numbness, “pins and needles”). A detailed neurologic examination including responses to vibration and pin-prick as well as deep tendon reflexes was performed in every patient. If neurologic symptoms or signs developed, nerve conduction studies (electromyography [EMG]) were performed.

The doses of thalidomide and dose modifications during treatment were obtained from the clinic notes. The patient's local pharmacy was the supplier of the drug. An informed consent was obtained for each patient. Patient and family were fully informed of the risk of teratogenicity, peripheral neuropathy, and sedation, and enrolled in the S.T.E.P.S. program (System of Thalidomide Education and Prescribing Safety, Celgene Corporation, Summit, NJ). Thalidomide is approved for marketing only under this special restricted distribution program (S.T.E.P.S.), approved by the Food and Drug Administration. The program is used for controlling access to thalidomide, educating prescribers, pharmacists, and patients, and monitoring compliance (2).

Female patients of childbearing potential were required to use 2 methods of contraception (1 greatly effective such as Depo-Provera and 1 barrier method) starting 4 weeks before thalidomide until 4 weeks after discontinuing thalidomide. Men were instructed to use latex condoms 4 weeks before starting thalidomide at the time of heterosexual intercourse and 4 weeks after discontinuing thalidomide. Female patients with childbearing potential had pregnancy tests performed 24 hours before starting thalidomide, then once weekly for 4 weeks and subsequently once every 4 weeks if they had regular menstrual cycles, and once every 2 weeks if irregular cycles.

Based on studies in adults, thalidomide was administered at bedtime, at a starting dose of 50 mg and increased stepwise by 50 mg increments to a maximum dose of 150 mg (mean 1.44 mg · kg−1 · day−1 [0.7–3.0 mg · kg−1 · day−1], if needed for persisting symptoms) according to patient's weight, symptoms, and adverse effects. Other immunomodulatory agents (azathioprine/6-mercaptopurine, methotrexate, anti-TNF-α biologics) were discontinued and not given concomitantly with thalidomide. In the event of an adverse reaction, the dose was reduced, discontinued, or withheld until the adverse effect had resolved.

The Harvey-Bradshaw Index (HBI) (20) was used to evaluate clinical response and remission. Clinical response for luminal disease was defined as a reduction in symptoms at 2 to 6 months on thalidomide and an HBI of 5 to 7. Clinical remission was defined as resolution of symptoms and a HBI <5. For fistulizing disease, clinical response was defined as a decrease of at least 50% in the number of draining fistulas. Clinical remission was defined as complete fistula closure.

All of the patients who received thalidomide were included in the analysis. Quantitative variables are described as medians and/or means with ranges in parenthesis throughout the text. The results were compared using a paired Student t test. Statistical significance level was set at P < 0.05 (2-sided).

Back to Top | Article Outline


Twelve patients were identified, 3 of whom were girls and 9 boys. Mean age at diagnosis was 10 years (range 3–14 years). Mean age of initiation of thalidomide therapy was 15 years (range 8–19 years).

Eight children had disease localized to the ileum and colon, 4 to the gastroduodenal area and colon, and 4 of 12 patients also had perianal disease. Five cases were complicated by strictures and 7 by fistulae. The mean number of anti-TNF infusions before thalidomide was 7 (range 1–21 infusions). All of the patients had failed anti-TNF biologic treatment. All 7 patients with fistulae received intravenous and oral antibiotics (metronidazole and/or ciprofloxacin) before starting thalidomide therapy without significant improvement. Five of 7 patients with fistulae continued oral antibiotics concomitant with thalidomide, whereas 2 received thalidomide alone. The characteristics of the patient population are listed in Table 1. Mean duration of thalidomide use was 39.5 months (1–96 months). The mean observation period before thalidomide use was 5.8 years, and mean follow-up after thalidomide was 4.6 years (P = 0.8).

The HBI, prednisone dose, number of hospitalizations, number of bowel resections, number of incision and drainage procedures, erythrocyte sedimentation rate, albumin, platelets, hemoglobin, and C-reactive protein were calculated before and after a 1- to 6-month (median 2 months) course of thalidomide (Table 2 and Figs. 1–3). During this period, the HBI scores decreased to <5 in 10 of 12 (83.3%) patients following thalidomide intervention, indicating clinical remission. HBI in 2 patients was consistent with response, but remained >5 (see Fig. 1). Of the 7 patients with fistulae, 5 had complete fistula closure (71.4%), 1 had partial closure (14.2%), and 1 had no improvement (14.2%) on thalidomide. The mean prednisone dose decreased from 13.9 to 2.3 mg/day. Seven of 12 patients were able to discontinue steroids (58.3%), 3 of 12 patients (25%) decreased their steroid dose by >50%, and 1 of 12 patients (8.3%) had no change in prednisone dose (Fig. 2).









During the follow-up period, on average 4.6 years, we observed a decrease in the frequency of surgical procedures performed. Before thalidomide, 6 of 12 (50%) patients had undergone intestinal resections, ranging from 1 to 4 bowel resections. The mean time of resections from diagnosis of CD before thalidomide was 24 months (range 3–96 months) After thalidomide, only 2 of 12 (16.6%) patients required bowel resections.

Five of 12 (41.6%) patients had incision and drainage procedures before thalidomide. The mean time of incision and drainage from diagnosis of CD before thalidomide was 12 months (range 7–24 months). Three of 5 patients had required 2 to 3 incision and drainage procedures. None underwent incision and drainage while on thalidomide. Two patients required seton placement before thalidomide without significant response and the setons were removed before thalidomide therapy.

The surgery rate was calculated for the patients before thalidomide and on thalidomide. The surgery rate prethalidomide (total of 26 surgeries [resections, incision and drainage, and setons] within 826 patient months) was 0.031. The surgery rate on thalidomide (total of 2 surgeries within 473.5 patient months) was 0.004.

The laboratory evaluations also improved: The erythrocyte sedimentation rate in 10 of 12 (83.3%) patients on thalidomide decreased to <20 mm/h (normal value <20 mm/h, no sex difference; Fig. 3). Similar statistically significant improvement was noted for hemoglobin, platelets, albumin, and C-reactive protein (Table 2).

Back to Top | Article Outline


Nerve conduction studies were obtained in 5 of 12 patients. Results of nerve conduction studies in 4 of 5 patients were consistent with mild sensory neuropathy and 1 of 5 had moderate sensory neuropathy. One patient with mild sensory neuropathy had documented resolution on EMG studies, after dose reduction of thalidomide from 50 mg daily to 50 mg once per week, and he remains on thalidomide. Clinical symptoms and findings of peripheral neuropathy resolved in all of the patients within 2 to 3 months after discontinuation (n = 4) or dose reduction (n = 2) of thalidomide. The mean cumulative dose of thalidomide in all of the patients was 51.7 g (1.5–166.0 g). In the patients who developed peripheral neuropathy, the mean cumulative dose of thalidomide was 54.4 g (16.5–102.8 g) versus 49.0 g (1.5–166.0 g) in the patients who had no peripheral neuropathy (P = 0.85).

Adverse reactions resulting in the discontinuation of thalidomide were peripheral neuropathy in 42% (5/12), worsening of the CD in 17% (2/12), dizziness in 8% (1/12), and an allergic reaction (swollen eyes) in 8% (1/12). Three of 12 patients (25%) remain on thalidomide; their mean time on thalidomide was 72 months (range 48–96 months.). Of the patients who continue on thalidomide, 1 had peripheral neuropathy, which resolved after dose reduction (Table 3). The patients who subsequently discontinued thalidomide after improvement in their CD were transitioned to other immunosuppression agents (Table 4).





Back to Top | Article Outline


In our study, we found significant improvement in all of the parameters measured in these pediatric patients with severe refractory CD treated with thalidomide, and that the effect was sustained for a prolonged period. We describe the longest follow-up (4.6 years) in comparison with previous adult and pediatric studies (14–16,18,19,21).

Our results suggest that pediatric patients with severe refractory CD may have better clinical response and remission rates than adult patients. In previous studies in adult patients with CD, clinical response ranged from 56% to 70%, clinical remission from 20% to 33%, and the fistula response from 33% to 83% (14,15,17,18,21). In comparison, our results in pediatric patients treated with thalidomide were 83.3% clinical remission and 71.4% complete fistula closure.

Lenalidomide, an analogue of thalidomide with similar immunomodulatory effects but 1000-fold greater potency and lacking the teratogenicity effects of thalidomide, has been also studied in CD. In contrast to thalidomide, it was not effective in adults with active CD (22).

Our study differs from a report of thalidomide use in Italian children with CD in that 100% of our patients failed anti-TNF biologics (infliximab or adalimumab) compared with 18% in their study (19). It is also the first investigation to measure fistulizing CD, hospitalizations, and surgeries in pediatric patients on thalidomide.

The main adverse effect of thalidomide is peripheral neuropathy, which was seen in 50% of our patients. The peripheral neuropathy is predominantly sensory with axonal degeneration. In previous studies, the reported incidence varied from 1% to 70% (23–25). A previous report of thalidomide use in children noted that neuropathy, if present, occurred 9 to 18 months after starting thalidomide therapy. The dose range associated with neuropathy was 2.3 to 10 mg · kg−1 · day−1 or cumulative doses of 55 to 1435 g, but peripheral neuropathy has been observed with cumulative dose of only 2.8 g. (24). The presenting symptoms of sensory conduction disturbance are numbness, paresthesias, hyperaesthesia, “pins and needles” sensation, and leg cramps. Recovery is variable and often incomplete. In children, paresthesias may resolve within months after cessation of treatment, but significant recovery from severe thalidomide neuropathy appears uncommon in childhood (24). In our study, the mean cumulative doses in the patients who developed peripheral neuropathy was 54.4 g compared with 49.0 g in patients with no clinical signs of neuropathy, which is consistent with previous studies. One of our patients had full recovery of neuropathy by EMG after a dose reduction.

Our study had some limitations. It is a retrospective uncontrolled evaluation involving a small number of patients. Although impaired sensory nerve conduction is a potential complication of thalidomide therapy, EMG studies were not performed in patients without symptoms or signs of sensory impairment because of the discomfort associated with these tests. Our estimation of reduction of surgical procedures could be overestimated by the duration of follow-up, on average, 4.6 years.

Because of the small number of patients who have received thalidomide therapy, optimal dosing of thalidomide is unknown for adult or pediatric patients. We used low-dose thalidomide (50 mg/day), increasing gradually to 150 mg in adolescents if clinically indicated for active CD. In this small series, we observed clinical and fistulae remission rates in the majority of pediatric patients with severe refractory CD.

In conclusion, thalidomide is a potentially effective drug for severe, refractory, and fistulizing CD in pediatric patients who have failed anti-TNF-α therapy. Our results showed significant improvement in laboratory measurements, hospitalization rate, reduction in the daily corticosteroid dose, and fistula closure. The main side effect was peripheral neuropathy observed in 50% (6/12) of the patients, which resolved clinically in all of the patients.

We anticipate that the results of a randomized controlled trial, under way in Europe, will provide additional information concerning the role of thalidomide in the treatment of pediatric patients with refractory CD.

Back to Top | Article Outline


The authors thank the Barnett family for their support in the Pediatric Inflammatory Bowel Disease Program at the University of Chicago.

Back to Top | Article Outline


1. Ginburg PM, Dassopoulos T, Ehrenpreis ED, et al. Thalidomide treatment for refractory Crohn's disease: a review of the history, pharmacological mechanisms and clinical literature. Ann Med 2001; 33:516–525.
2. Zeldin JB, Williams BA, Thomas SD, et al. S.T.E.P. S.: a comprehensive program for controlling and monitoring access to thalidomide. Clin Ther 1999; 21:319–330.
3. Sheskin J. Further observation with thalidomide in lepra reactions. Lepr Rev 1965; 36:183–187.
4. Sampaio EP, Sarno EN, Galilly R, et al. Thalidomide selectively inhibits tumor necrosis factor alpha production by stimulated human monocytes. J Exp Med 1991; 173:699–703.
5. Gordon JN, Goggin PM. Thalidomide and its derivatives: emerging from the wilderness. Postgrad Med J 2003; 79:127–132.
6. Westein TA, Sciubba JJ, Levine J, et al. Thalidomide for the treatment of oral aphthous ulcers in Crohn's disease. J Pediatr Gastroenterol Nutr 1999;28:214–6.
7. Bousvaros A, Mueller B. Thalidomide in gastrointestinal disorders. Drugs 2001; 61:777–787.
8. Hegarty A, Hodgson T, Porter S. Thalidomide for treatment of recalcitrant oral Crohn's disease and orofacial granulomatosis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003; 95:576–585.
9. Kolivras A, De Maubeuge J, Andre J, et al. Thalidomide in refractory vulvar ulcerations associated with Crohn's disease. Dermatology 2003; 206:381–383.
10. Yasui K, Uchida N, Akazawa Y, et al. Thalidomide for treatment of intestinal involvement of juvenile-onset Behcet disease. Inflamm Bowel Dis 2008; 14:396–400.
11. Rafiee P, Stein DJ, Nelson VM, et al. Thalidomide inhibits inflammatory and angiogenic activation of human intestinal microvascular endothelial cells (HIMEC). Am J Physiol Gastrointest Liver Physiol 2010; 298:G167–G176.
12. Prehn JL, Landers C, Muller GW, et al. Potent inhibition of cytokine production from intestinal lamina propia T cells by phosphodiesterase-4 inhibitory thalidomide analogues. J Clin Immunol 2001; 21:357–364.
13. Bauditz J, Wedel S, Lochs H. Thalidomide reduces tumor necrosis factor alpha and interleukin 12 production in patients with chronic active Crohn's disease. Gut 2002; 50:196–200.
14. Ehrenpreis ED, Kane SV, Cohen LB, et al. Thalidomide therapy for patients with refractory Crohn's disease: an open-label trial. Gastroenterology 1999; 117:1271–1277.
15. Vasiliauskas EA, Kam LY, Abreu-Martin MT, et al. An open label pilot study of low-dose thalidomide in chronically active, steroid-dependent Crohn's disease. Gastroenterology 1999; 117:1278–1287.
16. Facchini S, Candusso M, Martelossi S, et al. Efficacy of long-term treatment with thalidomide in children and young adults with Crohn's disease: preliminary results. J Pediatr Gastroenterol Nutr 2001; 32:178–181.
17. Kane S, Stone LJ, Ehrenpreis E. Thalidomide as “salvage” therapy for patients with delayed hypersensitivity response to infliximab. J Clin Gastroenterol 2002; 35:149–150.
18. Sebate JM, Villarejo J, Lemann M, et al. An open-label study of thalidomide for maintenance therapy in responders to infliximab in chronically active and fistulizing refractory Crohn's disease. Aliment Pharmacol Ther 2002; 16:1117–1124.
19. Lazzerini M, Martelossi S, Marchetti F, et al. Efficacy and safety of thalidomide in children and young adults with intractable inflammatory bowel disease: long-term results. Aliment Pharmacol Ther 2006; 25:419–427.
20. Harvey RF, Bradshaw JM. A simple index of Crohn's disease activity. Lancet 1980; 1:514.
21. Plamondon S, N.G. SC, Kamm MA. Thalidomide in luminal and fistulizing Crohn's disease resistant to standard therapies. Aliment Pharmacol Ther 2006; 25:557–567.
22. Mansfield JC, Parkes M, Hawthorne AB, et al. A randomized, double-blind, placebo-controlled trial of lenalidomide in the treatment of moderately severe active Crohn's disease. Aliment Pharmacol Ther 2007; 26:421–430.
23. Strauss RS, Das KM. Thalidomide-induced sensory neuropathy. J Pediatr Gastroenterol Nutr 2001; 32:322–324.
24. Fleming FJ, Vytopil M, Chaitow J, et al. Thalidomide neuropathy in childhood. Neuromuscul Disord 2005; 15:172–176.
25. Bastuji-Garin S, Ochonisky S, Bouche P, et al. Incidence and risk factors for thalidomide neuropathy: a prospective study of 135 dermatologic patients. J Invest Dermatol 2002; 119:1020–1026.

anti-TNF biologics; Crohn disease; peripheral neuropathy; thalidomide

Copyright 2012 by ESPGHAN and NASPGHAN