A Phase II Dose Titration Study of Thalidomide for Cancer-Associated Anorexia

Abstract

Context

Sixty-five percent of people with advanced cancer suffer from loss of appetite. Several inflammatory cytokines appear to cause appetite loss in animal models. Thalidomide is an immunomodulatory drug that has been associated with improved appetite in those with HIV infections, and in cancer.

Objectives

We completed a two-stage Phase II dose titration study of thalidomide, the primary purpose of which was to assess appetite response to thalidomide in cancer-associated anorexia.

Methods

Individuals older than 18 years of age with active cancer, loss of appetite by numerical rating scale (NRS), life expectancy of at least four weeks, and Eastern Cooperative Oncology Group performance status of 0–3 were entered into the study. Pre-treatment screening included medical history, neurologic examination, and symptoms by NRS and categorical scale. Patients received 50 mg of thalidomide by mouth at bedtime for two weeks. Individuals who did not respond were dose escalated to 100 mg at night for two weeks. Assessment of appetite, early satiety, fatigue, insomnia, night sweats, pain, and quality of life occurred at two-week intervals. Toxicity also was assessed. The primary outcome was appetite response defined as a two-point reduction in the NRS, or one-point improvement in the categorical scale.

Results

Thirty-five patients entered the study; 33 completed 14 days of therapy and were analyzed for efficacy and toxicity. Sixty-four percent who completed at least two weeks of thalidomide had improved appetite. The categorical scale scores for appetite, insomnia and quality of life improved significantly. The 95% confidence intervals did not overlap. Five participants dropped out because of toxicity: two before two weeks and three later.

Conclusion

Thalidomide reduced multiple symptoms commonly associated with cancer-related anorexia and improved quality of life. Our findings confirmed and validated a previously published single arm trial. A recent randomized trial demonstrated greater benefits when thalidomide is used combined with other agents to treat cancer-associated anorexia and cachexia. Thalidomide helped cancer-associated anorexia in a majority of patients. It also improved insomnia and quality of life in advanced cancer.

Introduction

Sixty-five percent of advanced cancer patients suffer from anorexia and over half will have unintended weight loss greater than 10% of their pre-morbid body weight (1). Most who report anorexia will have concurrent gastrointestinal symptoms like bloating, constipation, diurnal changes in food intake, early satiety and an altered sense of smell and taste (2). The presence of anorexia and significant involuntary weight loss portends a short survival and high symptom burden (3). Anorexia is part of a symptom cluster along with weakness, fatigue and lack of energy. Treatment of one symptom within this cluster might also ameliorate other cluster symptoms (4). Effective management of appetite loss and weight loss should reduce symptom burden, and perhaps favorably influence survival.

The development of a validated instrument to accurately measure anorexia is important to research and clinical management. We recently validated a two-item instrument that measured appetite on a numerical rating scale (NRS) (0 is complete loss of appetite and 10 is normal appetite) and patient-perceived severity by categorical scale (CAT) (5). This simplified questionnaire had a high concurrent and predictive validity when compared to the validated Functional Assessment of Anorexia and Cachexia Therapy shortened 12 Question Version (5). It had moderate sensitivity to changes in appetite over seven days when compared to the validated scale.

Mechanisms of appetite loss and unintended weight loss are protean. Several inflammatory cytokines appear to cause anorexia and cachexia in animal models (6). Serum levels of Interleukin-1 and 6, tumor necrosis factor (TNF) alpha, vascular endothelial growth factor (VEGF)-A and VEGF-C are elevated in cancer, and associated with weight loss (7). Systemic inflammation-based prognostic scales, like the Glasgow Prognostic Score (GPS), predict nutritional decline and prognosis independent of cancer stage (8). This provides a well-defined therapeutic target for clinical trials (8).

Thalidomide is an immunomodulatory drug that inhibits VEGF and TNF alpha (9). It was originally licensed for erythema nodosum leprosum in 1997. However, studies have shown thalidomide reduces wasting associated with HIV infections (9). Significant weight gain was seen with thalidomide 100 mg daily when administered to those with wasting from HIV infection. However, over half of the participants discontinued because of adverse events including somnolence, rash, neutropenia, neuropathy and dizziness (9). In another small randomized trial of 28 individuals with HIV-associated wasting, weight gain was greater on low thalidomide doses compared to placebo; the Karnofsky Index also improved relative to placebo (10). Somnolence and rash were more common with thalidomide. Response was seen in 63% (22 of 35) of evaluable patients with cancer-associated cachexia. Insomnia, nausea and well-being also improved (11). Reported toxicity was mainly drowsiness and dizziness.

We completed a thalidomide dose titration study. The primary purpose was to assess the appetite response to thalidomide in cancer-associated anorexia. We also evaluated safety. Secondarily, we wanted to determine if weight gain occurred with thalidomide and if thalidomide influenced performance score measured by the Eastern Cooperative Oncology Group (ECOG) performance score. Changes in lean body mass (LBM) measured by bioelectric impedance (BIA), fatigue, early satiety, insomnia, pain and quality of life were secondary outcomes.

Methods

Twenty individuals (minimum) and 32 individuals (maximum) with advanced cancer were to be enrolled in this dose titration study. Patients were recruited from referrals to the Palliative Medicine Service, and from the Department of Solid Tumor Oncology within the Cleveland Clinic Taussig Cancer Institute. Inclusion criteria were age greater than 18 years, active cancer, anorexia by NRS (0 no appetite, 10 normal appetite), life expectancy of at least four weeks, ECOG performance s of 0–3, and ability to understand and communicate in English. This was an institutional review board-approved study. Patients had to be willing and able to comply with the Food and Drug Administration (FDA) mandated S.T.E.P.S. (System for Thalidomide Education and Prescribing Safety) program before study entry. Females of child bearing potential who refused to use two methods of birth control (to start four weeks before thalidomide) were excluded. Male patients with female partners of childbearing potential were excluded if they refused to wear latex condoms during sexual encounters. Other exclusion criteria were: peripheral neuropathy greater than 2 as published by Feldman and Stevens (12), concurrent chemotherapy or radiation, hematologic malignancies, those unable to take oral medications and food or who required tube feedings or parenteral nutrition, and use of appetite stimulants like corticosteroids, dronabinol, growth hormone, megestrol acetate, oxandrolone, or pentoxifylline. Those individuals who, in the investigator’s opinion, were at unacceptable risk if they participated also were excluded.

Study treatment consisted of thalidomide 50 mg administered by mouth at bedtime for two weeks. Those who responded were maintained on thalidomide for six weeks at the investigator's discretion, and patient's wish to continue therapy. Individuals who did not respond to 50 mg and were without dose-limiting toxicity, were dose escalated to 100 mg at night for two weeks; response was then re-assessed. Patients who had not responded to 100 mg and did not have dose-limiting toxicity could be dose escalated to 200 mg at bedtime. Those who experienced severe or serious adverse events were withdrawn from the study. Adverse events that could lead to discontinuation were: unacceptable sedation and drowsiness as judged by the patient or the physician, peripheral neuropathy, development of an exfoliative, purpuric or bullous rash, fever > 38°C documented on clinical measurements separated by 24 hours, or seizures. Other side effects, which also were monitored by CAT (0 absent, 1 mild, 2 moderate, 3 severe), included constipation, confusion, pruritus, pain and swelling in the limbs or face. Treatment was discontinued if side effects were deemed severe or serious by the treating physician.

Thalidomide was obtained from Celgene Pharmaceuticals (Destin, Florida) after patients enrolled in the S.T.E.P.S program. Thalidomide 50 mg capsules contained microcrystalline cellulose NF (60 mg), povidone 90 F USP (12 mg), stearic acid NF (4 mg), colloidal silicon dioxide NF (0.8 mg), crospovidone NF (16 mg) and anhydrous lactose NF (257.2 mg). Patients were instructed to take their dose at bedtime.

Assessment of Response and Toxicity

Pre-treatment screening included the following: medical history, neurologic examination, appetite by NRS (0 = no appetite, 10 = normal appetite) and CAT (0 = normal appetite; level 1= 75% of normal intake; level 2 = 50% of normal intake; 3 = < 50% of normal intake; 4 = not eating at all), strength by NRS (0 = no strength, 10 = normal strength), pain by NRS (0 = no pain, 10 = severe pain), insomnia by NRS (0 = sleeping well, 10 = not sleeping at all), sweating by NRS (0 = no sweating, 10 = severe sweating), nausea by NRS (0 = no nausea, 10 = severe nausea), tiredness by CAT (0 = no tiredness, 5 = cannot take care of myself), early satiety by CAT (0 = normal sense fullness after meals, 3 = always full), and quality of life by a single-item CAT (0 = quality of life very bad, 4 = very good). Toxicities were assessed using a CAT (absent = 0, mild = 1, moderate = 2, severe = 3), not Common Toxicity Criteria and included pain, muscle cramps, confusion, constipation, dizziness, drowsiness, dry mouth, headache, loss of interest in sex, nervousness, pain in the stomach, dry skin or pruritus, and limb swelling. Performance score was measured by the ECOG performance scale (higher score worse). Weight was measured in kg; LBM by BIA, C-reactive protein (CRP) by standard, not highly sensitive CRP. Pregnancy tests were done according to the S.T.E.P.S. protocol. Patients underwent a weekly history and physical examination and were assessed for response and toxicity by telephone every 48 hours. After six weeks, individuals were followed in the outpatient clinic every two weeks.

The primary outcome was appetite response defined as a two-point increase in NRS without severe or serious adverse events. Secondary endpoints were score changes in performance status (one point), pain (two points), insomnia (two points), nausea (two points), night sweats (two points), fatigue (one point), early satiety (one point) and quality of life (one point). Higher scores were worse for all except quality of life. Compliance monitoring was not performed.

Statistical Considerations

Descriptive statistics summarized variables at each time point of assessment. Categorical variables were summarized as frequencies and percentages. Analysis of efficacy was based on those given at least two weeks of thalidomide. Eligibility for assessment of toxicity was based on receipt of a single dose of thalidomide. Treatment success was established if 60% of treated individuals had improved appetite. Failure was a response rate of less than or equal to 40%. Twenty patients were initially entered into the study. If eight or less of these responded, the study was to be stopped. If nine or more responded, an additional 12 would be added. If 16 individuals or less responded over the two-week trial period, the study would be considered negative. The power for the primary outcome measure was 0.83, with a significance level of 0.09.

Changes in symptom severity from baseline were analyzed by the Wilcoxon W test. Comparison of responders to non-responders was made with the Mann-Whitney test. Variables were compared using the Wilcoxon W test. Categorical variables were compared using either Fisher’s exact test or the McNemar test. Confidence intervals (CI) from the standard error of the means were calculated for all significant P-values. All statistical tests were two-sided and the significance level was P < 0.05.

Results

Thirty-five patients gave consent and were enrolled in the study. Two individuals did not complete two weeks of therapy and were excluded from analysis; 33 patients completed at least 14 days of therapy and were evaluable for both response and toxicity. Demographics of the patient population are shown in Table 1. On day fourteen, 17 of 33 individuals had improved appetite. Nine of the 16 non-responders were then titrated to 100 mg of thalidomide; seven of these withdrew after two weeks (four refused titration, and three had severe side effects as described below). By day 21, four of the remaining nine individuals had responded; five were non-responders. Sixty-four percent of individuals who completed at least two weeks of thalidomide had improved appetite. In addition early satiety, insomnia, pain and quality of life improved significantly (Table 2).

Table 1.

Demographic Data (n = 33)

		Mean (SD)	Median (Range)
Age, yrs		68 (13)	69 (43–87)
		n	%
Gender
	Male	17	51%
ECOG score
	0	1	3%
	1	15	47%
	2	14	41%
	3	3	9%
	4	0	0
Diagnosis
	Lung Cancer	13	39%
	GI Cancer	12	36%
	Breast Cancer	2	6%
	Head & Neck	2	6%
	Others	4	12%

All numbers rounded to the nearest whole number.

Table 2.

Mean Symptom Scores at Baseline and Days 14–21 (n=33)

Symptom	Baseline		Day 14 or 21		P-valuea
	Mean (SD)	(95%CI)	Mean SD)	(95%CI)
Appetiteb (by NRS)	3.5 +/− 2.0	(2.8 – 4.2)	5.7 +/− 2.6	(4.8 – 6.5)	< 0.001
Appetitec (by CAT)	2.5 +/− 0.6	(2.3 – 2.7)	1.9 +/− 0.9	(1.6 – 2.2)	< 0.001
Other Symptomsc
Sweating	0.1 +/− 0.3	NS	0.5 +/− 1.7	NS	NS
Pain	1.7 +/− 2.5	(0.9 – 2.6)	0.7 +/− 1.5	(0.2 – 1.2)	<0.05
Night Sleep	2.5 +/− 3.0	(1.5 – 3.5)	0.7 +/− 1.5	(0.2 – 1.2)	< 0.01
Nausea	0.5 +/− 1.7	NS	0.6 +/− 1.6	NS	NS
Tired	2.7 +/− 1.3	NS	2.7 +/− 1.4	NS	NS
Early Satiety	1.2 +/− 1.1	(0.8 – 1.6)	0.8 +/− 1.0	(0.5 – 1.1)	< 0.05
Quality of Life
QOL (CAT)	1.2 +/− 0.7	(0.5 – 0.9)	1.7 +/− 0.7	(1.4 – 1.9)	<0.01

NS = non-significant P-values (CI only calculated for significant P-values).

^aWilcoxon W.

^bHigher scores better.

^cHigher scores worse.

Side effects were negligible for most (Table 3). The mean score of the 13 potential side effect symptoms monitored by CAT did not increase significantly for the entire group. Two individuals did drop out because of side effects before completing 14 days of therapy and were not reflected in the mean toxicity score. Three withdrew after 14 days because of side effects: one from severe constipation, another from severe constipation and pruritus, and one with severe dry mouth.

Table 3.

Side Effects Associated with Thalidomide

Side Effects	Baseline a n, mean (SD)	Thalidomide a n, mean (SD)	P-value
Peripheral Neuropathy	4, 0.2 (0.5)	3, 0.1 (0.4)	NS
Drowsy	13, 0.5 (0.7)	5, 0.2 (0.6)	< 0.05 b
Confusion	7, 0.3 (0.7)	4, 0.3 (0.7)	NS
Nervousness	8, 0.3 (0.7)	5, 0.2 (0.5)	NS
Loss of Libido	6, 0.9 (1.3) (15 missing)	1, 0.2 (0.7) (18 missing)	= 0.1
Headache	2, 0.1 (0.2)	0, 0 (0)	NS
Dry Mouth	15, 1.3 (1.4)	16, 1.1 (1.3)	NS
Abdominal Pain	5, 0.3 (0.7)	3, 0.2 (0.5)	= 0.08
Constipation	11, 0.8 (1.2)	13, 0.8 (1.1)	NS
Itching	3, 0.3 (0.8)	3, 0.3 (0.9)	NS
Limb/Face Swelling	6, 0.3 (0.8)	6, 0.3 (0.6)	NS
Dizzy	7, 0.4 (0.8)	5, 0.3 (0.6)	NS
Skin Rash	0	1	NS
Fever	0	0	NS
Edema	5, 0.3 (0.9)	6, 0.6 (1.3)	NS

NS = non-significant P-values.

^aHigher mean scores worse.

^bWilcoxon W.

Discussion

Thalidomide has a wide spectrum of actions including anti-inflammatory, anti-angiogenic and anti-emetic activity, sedation, and immunomodulation (13). The anti-inflammatory and immunomodulatory effects are largely attributable to down-regulation of TNF alpha messenger RNA (14). The anti-cachectic action is closely related to modulation of TNF alpha (15). TNF alpha and Interleukin-1 up-regulate the hypothalamic melanocortin system, and down-regulate neuropeptide Y, causing anorexia (16). Down-regulation of TNF alpha would improve appetite in inflammatory disorders like HIV infections and cancer (18–20).

We found that thalidomide improved appetite, sleep, and quality of life to a significant degree, and reduced early satiety and pain but less so (demonstrated by overlapping confidence intervals for outcome measures means). Four other studies investigated thalidomide to treat cancer anorexia and/or cachexia (Table 4). Two were randomized controlled trials, one of which was double-blind and compared thalidomide to placebo (21). This smaller trial, though double-blind, had a high dropout rate at the first assessment (17 of 47 individuals). LBM as measured by mid-arm circumference improved at two weeks (21). Weight also improved. There was no change in handgrip strength (a surrogate marker for fatigue and/or muscle mass). In the second trial, thalidomide was compared to various anti-cachexins (22). This study demonstrated no appetite benefit by visual analogue scale (22). There was no improvement in LBM (by bioelectrical impedance or dual-emission X-ray absorptiometry [DXA] scans), fatigue, resting energy expenditures, handgrip, or appetite over the four months on thalidomide alone. Interleukin-1 and GPS scores improved with thalidomide (22). Global health scores did not improve in either study (21, 22). Side effects were negligible in the second trial, whereas the first reported neuropathy, skin rash, somnolence, improved insomnia and constipation in some (21).

Table 4.

Comparison of Thalidomide Studies

Study	#	Design	Dose	Outcomes	Attrition	Results	Toxicity
Mantovani et al.22	332 (87 on thalidomide alone, 88 on combination with thalidomide)	RCT, not double-blind 4-month duration	200 mg/d	LBM, REE, appetite, handgrip, QOL, BIA, LBM, IL6, TNF∂, GPS, fatigue	3/87 (single thalidomide arm of the trial)	↓ ECOG-PS, ↓IL-6 ↓GPS No improvement in appetite fatigue, LBM, REE, handgrip	Negligible
Khan et al.23	11	Single arm GI cancers 2 weeks	200 mg	LBM, REE, Karnofsky Index	1/10	↑ Weight, ↑ LBM, ↑ QOL No improvement in REE	N/A
Gordon et al.21	47	RCT / double-blind Pancreatic cancer 4 weeks	200 mg	Weight Δ, muscle mass (AMA), handgrip, survival, global health score	17/47 at 4 weeks, 27/47 at 8 weeks	↑ Weight gain and maintained LBM. No change in handgrip, global health score, survival	Neuropathy Rash Somnolence Constipation ↓Insomnia
Bruera et al.11	37	Single arm 10 days	100 mg	Difficulty falling asleep, restlessness in AM, insomnia, nausea, appetite, well-being	35/72	↑ Insomnia ↑appetite ↓ nausea ↑ well-being	Dizziness Drowsiness Dry mouth Lightheadedness
Davis et al. (present study)	33	Single arm All cancers 2 weeks	50 mg/100 mg	Appetite, weight loss, nausea, night sweats, fatigue, early satiety, pain, insomnia	2/35 (14 days)	↑ Appetite ↓ insomnia ↓early satiety ↓ pain ↑ QOL No improvement in weight, ECOG PS, nausea, fatigue	Negligible by CAT

In three single-arm studies (to include our study), 83 patients were treated (23, 24). Doses were 50 mg to 200 mg. A small study treated 11 individuals with esophageal cancer. Compared to weight loss during a two-week, before treatment, run-in period, thalidomide reduced the rate of weight loss after two weeks of 200 mg per day (23). LBM by DXA scan increased with thalidomide. Quality of life by Karnofsky Index also improved. Basal metabolic rate was reduced, but the differences in the means had overlapping confidence intervals. Mild sedation was experienced by most but disappeared after two to three days (23).

In the larger single arm study, 37 of 72 individuals who started the study were evaluable (11). Initial thalidomide doses were 100 mg a day and responses were assessed approximately 10 days after initiating therapy. A severity NRS measured difficulty falling asleep, restedness in the morning, insomnia, nausea, appetite and well-being as primary outcomes. A 30% NRS reduction was deemed a response. Demographic characteristics resembled our study. Appetite response was identical. In addition, insomnia and well-being improved in over 50% of treated individuals. Nausea improved in 44%. Most chose to continue thalidomide after the ten-day study. Dizziness, drowsiness and dry throat led to discontinuation in five individuals. Most adverse effects were mild (11).

Two studies of similar design (which includes our own) demonstrated improved appetite with thalidomide (11). Our experience validates the earlier study. The fact that the percentage response was the same, and that statistical significance occurred without overlapping confidence intervals by CAT measures (as in our study) provides confirmatory evidence that thalidomide improves appetite in cancer. Contrary to the two single-arm studies, the large randomized trial did not find appetite improvement when thalidomide was added to a specialized polyphenol diet (22). Appetite was a secondary endpoint, and comparisons were made 16 weeks after starting thalidomide. Appetite did improve when an omega-3 fatty acid rich supplement, L-carnitine, and progesterone were added to thalidomide. It may be that the appetite benefits with thalidomide were lost over time, the specialized polyphenol diet (plus vitamins) may have stimulated appetite like thalidomide, or that there are additive appetite benefits with combinations of agents.

Insomnia improved in both studies that utilized it as an outcome. Improved insomnia with thalidomide was reported over 40 years ago (24, 25) but not in cancer until recently. Insomnia is one of the major symptoms in advanced cancer (26) so a medication that improves appetite but also reduces insomnia is valuable.

The reported LBM improvement with thalidomide is controversial. The larger randomized trial (longer follow-up than the smaller studies) did not demonstrate improved LBM (22). In the smaller randomized trial that used anthropometric measurements, LBM was stable-to-improved on thalidomide at four and eight weeks; better than placebo (21). This study had a high dropout rate, an imprecise LBM method, and some sample and measurement biases. The other single-arm study demonstrated some LBM improvement by DXA scans but had a small sample size (23). Treatment time was two weeks. Sample bias is likely. Even if individuals are healthy and undergo resistance training to increase muscle mass, size and fiber cross-sectional area do not increase for eight weeks (8). It would be difficult to see improved muscle mass and LBM in such a short time in advanced cancer. Therefore, it is likely that the improved LBM in these two small trials reflected bias or sample variability and selection, and imprecise measurement. Thalidomide may prevent or forestall LBM loss; therefore, confirmatory evidence is needed to prove it improves with thalidomide.

Neither fatigue nor handgrip improved with thalidomide. Two studies demonstrated improved well-being and quality of life, and one did not. Quality-of-life instruments differed, and influences on quality of life may reflect different domains measured. This requires further investigation.

Side effects with thalidomide were mild, occurred in a minority in all studies and led to drug discontinuation in few individuals. This may be in part dose-dependent, since the highest dose was 200 mg. However, the analysis of most studies was based on the treatment given, and not intention-to-treat. The large randomized trial had the longest treatment and few dropouts; toxicity was negligible. Hence, most individuals tolerated thalidomide. The most common complaints were sedation, drowsiness and dizziness. Rash, constipation and neuropathy were rare. Most studies were less than four weeks in duration, which may have played a role in the reported side-effect profile. This may be particularly important with neuropathy (which usually requires several months to develop).

The GPS improved with thalidomide in the largest randomized control trials (22). Interleukin-1 also was reduced with thalidomide (22). The GPS is based on CRP and serum albumin. Appetite may not improve with thalidomide despite reduction in systemic inflammatory markers (8, 22). Other thalidomide mechanisms for appetite modulation need to be considered.

There were weaknesses in our study. Secondary outcomes involved CRP and BIA, which were not completed. A non-validated CAT was used for toxicity rather than the Common Terminology Criteria for Adverse Events (CTCAE). It was believed that the CTCAE would be inadequate to assess thalidomide toxicity. Pain and early satiety improved significantly (P-values) but confidence intervals overlapped, weakening the conclusion that improvements were from thalidomide. It was not possible to separate placebo from a drug benefit in this single-arm, open-label study. A placebo effect is known to be significant for pain; this is likely also true for anorexia. Changes in the mean symptom score (even if significant) can indicate either a large response in a few individuals, or little response in most. The percentage of those in our study who had a two-point change by NRS indicates that most benefited. Individuals on our study were on multiple other medications including opioids; pain could have improved as a result of analgesic titration, independent of thalidomide response. The same could be true for quality of life. This non-consecutive sample was small and may not represent the population of patients in our practice, although age, gender and distribution of cancer sites of those on study resembled our usual patient population. We excluded hematologic malignancies because thalidomide is front-line treatment for multiple myeloma (28). Symptom improvement in lymphoplasmacytoid disorders could be from tumor response rather than appetite stimulation. Toxicity was low but this may have been a result of the short treatment time and low thalidomide dose. These findings are hypothesis generating. Insomnia, early satiety, pain and quality of life were secondary outcomes. The study was not powered for secondary outcomes nor randomized.

Thalidomide should be investigated further in randomized studies both as a single agent and as part of combination therapy. Novel multiple endpoints should be considered for these studies given thalidomide’s apparent benefit in appetite, insomnia and quality of life. It cannot be recommended for routine clinical use based on our current information.

Conclusions

Thalidomide often improved appetite, insomnia and quality of life in advanced cancer. Toxicity was, in general, low. The therapeutic dose for symptom response is unknown: from our experience, responses occurred with low doses (50–100 mg). Future randomized studies should confirm these findings and explore various doses. Combination therapy (thalidomide with L-carnitine, progesterone, and omega-3 fatty acids) may hold promise (22).