Synergism between narrowband ultraviolet B phototherapy and etanercept for the treatment of plaque-type psoriasis


  • Funding source None.
  • Conflicts of interest None declared.


PierGiacomo Calzavara-Pinton.




Previous investigations have demonstrated that a combination of etanercept (ETN) and narrowband ultraviolet B (NB-UVB) phototherapy is more effective than ETN alone. However, it is unclear if this combination is more effective than NB-UVB phototherapy alone.


To evaluate whether the combination of NB-UVB phototherapy with ETN improves the efficacy of ETN alone in the treatment of moderate-to-severe psoriasis.


We enrolled 322 consecutive patients with moderate-to-severe plaque-type psoriasis, who were treated with NB-UVB phototherapy as the first-line treatment option. Patients who did not achieve a 75% improvement in Psoriasis Area and Severity Index (PASI 75) were treated with conventional systemic therapies for psoriasis. If they were ineligible for these, they were treated with ETN 50 mg twice weekly. If they did not achieve PASI 75 within 12 weeks, NB-UVB phototherapy was added.


PASI 75 was achieved in 262 patients (81·4%) treated with NB-UVB phototherapy. Sixteen patients (5·0%) dropped out for personal reasons and 24 (7·5%) were treated with at least one of the conventional systemic treatments for psoriasis. Twenty patients (6·2%) were treated with ETN. The combination regimen was needed in eight patients (2·5%) with poor response to both phototherapy and ETN alone. All of these patients achieved PASI 75 and three of them had a complete remission after 14·6 ± 3·3 NB-UVB exposures. The combined treatment was well tolerated without acute adverse events. Unfortunately, all of these patients relapsed, with PASI > 10 within 2·8 ± 1·7 months.


The combined treatment has a synergistic effect for clearing plaque-type psoriasis previously unresponsive to ETN and NB-UVB phototherapy alone. The clearance rate is very high in a very short time without short-term adverse effects. However, concerns regarding potential cocarcinogenicity remain. Therefore the number of patients who require, and could benefit from, the combined treatment is likely to be small.

Psoriasis is a chronic and recurrent inflammatory skin disorder. Activated lymphocytes and proinflammatory cytokines play pivotal roles in the development and maintenance of lesions. In order to selectively block the most relevant pathological pathways, biological drugs such as the tumour necrosis factor (TNF)-α inhibitors etanercept (ETN), infliximab and adalimumab; the antilymphocyte function-associated antigen-3 antibody alefacept; the anti-CD11a antibody efalizumab; and the anti-interleukin-12/23 antibody ustekinumab have been developed. Aiming at further improving their safety profile and efficacy, and possibly also to decrease the high cost of a treatment cycle, combinations with other antipsoriatic treatments have been explored.[1] Among these, narrowband ultraviolet B (NB-UVB) phototherapy appears attractive. NB-UVB has high efficacy, lacks systemic toxicity and is devoid of pharmacological interactions. A few clinical studies have determined that NB-UVB enhances the efficacy of commercially available biologics, including ETN,[2-5] adalimumab,[6] ustekinumab,[7] alefacept[8] and efalizumab,[9] with very good tolerability and without short-term adverse events. However, concerns about long-term adverse events, namely skin carcinogenesis, have been raised.

Furthermore, the clinical usefulness of such combination therapies has been challenged because there are no studies that assess whether – or to what extent – the combination with biological agents enhances (or decreases, or has no effect on) the efficacy of NB-UVB phototherapy alone. The obvious reason is that control lesions treated with phototherapy alone cannot be included in the experimental design of an intraindividual (e.g. half-side) comparative study. The comparison with historical data has considerable limitations.[10]

A possible approach to address this question is to assess the clinical efficacy of the combination treatment in patients who proved to be resistant to both NB-UVB phototherapy and a biological agent when used alone.

We report the treatment results of a group of 322 consecutive patients with psoriasis who were treated with NB-UVB as a first-line treatment approach, ETN as a second step in those 20 patients who were unresponsive to NB-UVB and, finally, the combination therapy of ETN plus NB-UVB as the third treatment option in the eight patients who were unresponsive to each therapy alone.

Patients and methods

Study design

In the first phase, aiming to identify those patients who were resistant to both NB-UVB and ETN therapies, the study was designed as a prospective clinical trial. All consecutive patients with moderate-to-severe psoriasis [Psoriasis Area and Severity Index (PASI) > 10] presenting to the Outpatient Clinic of the Brescia University Hospital were treated with NB-UVB phototherapy as a first treatment approach unless they had contraindications to the treatment[1] or they were unable to comply with the phototherapy regimen for personal reasons. Patients unresponsive to NB-UVB [< 75% PASI improvement (PASI 75) after 8 weeks of therapy] were considered for conventional systemic therapies, including methotrexate, ciclosporin, acitretin and psoralen plus ultraviolet A (PUVA) photochemotherapy.[1] If these treatments were contraindicated, had led to insufficient response (< 75% PASI improvement) or had not been tolerated in previous treatment cycles,[1] patients were eligible for treatment with ETN (Enbrel; Pfizer Italia SrL, Latina, Italy). Patients on ETN who did not achieve at least PASI 75 within 12 weeks were candidates to receive a combined treatment with ETN and NB-UVB phototherapy. At this point, the study design became a randomized controlled intraindividual comparison study, because two representative plaques of similar severity and size in the same body area were selected and randomized to be shielded or not with an opaque dressing before every NB-UVB treatment. Covered plaques thus served as nonirradiated controls. These plaques were evaluated at each visit with the Psoriasis Severity Index (PSI), which assesses the degree of three parameters – erythema, infiltration and scaling – on a five-point scale (0, absent; 1, slight; 2, moderate; 3, severe; 4, very severe) with a maximum score of 12.[9]

The combined treatment was delivered until clearance (PASI 100) of the exposed skin or until a partial improvement was achieved without further improvement with a further 2 weeks of combined treatment, with a minimum treatment time of 2 months. At this point, both ETN and phototherapy were stopped without a maintenance therapy and patients entered in a follow-up programme with monthly visits until recurrence of psoriasis with PASI ≥ 10. In order to avoid the influence of natural sun exposure on the course of the disease, the study was carried out from October 2010 to June 2011 and from October 2011 to June 2012.

The study was conducted in accordance with the Declaration of Helsinki and followed a protocol approved by our institutional ethics committee.

Patients and treatments

Patients to be enrolled were adults affected by moderate-to-severe plaque-type psoriasis (PASI score ≥ 10) without clinical signs or symptoms of joint involvement. Previous treatments for psoriasis were discontinued at least 2 months before entering the present study. Patients not eligible for NB-UVB phototherapy[1] and those who could not maintain a regular phototherapy regimen were excluded.

The initial dose was 0·1–0·4 J cm−2 according to skin phototype. Patients were treated three times each week (on Monday, Wednesday and Friday) and UV doses were adjusted at each session according to the erythema response. In short, 10%, 5% or 0% increments were delivered depending on the erythema response, if any, after 48 h.[11]

In the selection and management of the conventional therapeutic option for psoriasis, we adopted the exclusion criteria and restrictions on use from the European Dermatology Forum (EDF) and European Academy of Dermatology and Venereology (EADV) guidelines.[1] In addition, we used the recommended screening and monitoring guidelines.[1] Conventional systemic treatments for psoriasis were administered as follows: oral methotrexate (Pfizer Italia SrL) was given at a weekly dose of 15 mg, oral ciclosporin (Sandimmun Neoral; Novartis Italia SpA, Origgio, Italy) at a fixed daily dose of 200 mg, and oral acitretin (Roche SpA, Milan, Italy) at a daily dose of 1 mg kg−1. For PUVA photochemotherapy, 8-methoxypsoralen (Melaoline; Uni-Pharma SA, Kefisia, Greece) was given orally at a dose of 0·5 mg kg−1 followed by exposure to gradually increasing UVA doses. Treatment sessions were delivered three times per week.[1]

ETN was administered subcutaneously at a dose of 50 mg twice weekly. Before initiating ETN treatment patients underwent screening as suggested by the EDF/EADV guidelines[1] and patients meeting any exclusion criteria were not enrolled. Patients were monitored according to the same guidelines.

In the combined regimen, both phototherapy and ETN were delivered as described above. Throughout the study, patients were allowed to use only topical emollients in addition to NB-UVB and/or ETN therapy.

Radiation sources

NB-UVB radiation was delivered by 40 Philips TL-01/100 W fluorescent tubes (Philips, Eindhoeven, the Netherlands) housed in a Waldmann 7001K stand-up cubicle (Waldmann Lichttechnik GmbH, Villingen-Schwenningen, Germany).

UVA radiation was emitted by 40 Philips TL-12/100 W fluorescent tubes (Philips) housed in a stand-up PUVA Combilight cubicle (DELCOMp bvba, Leuven, Belgium). Irradiances were measured with an SR 9910 spectroradiometer (Macam Photometrics Ltd, Livingston, U.K.).

For statistical analysis all calculations were carried out according to an intention to treat analysis. Results are reported as mean ± SD. PASI and PSI scores were analysed using the Student t-test for paired samples. Significance was defined as < 0·05.


We evaluated 368 consecutive patients. Eleven patients were not enrolled in the study due to contraindications to phototherapy. An additional 35 patients did not agree to undergo phototherapy because of travel time, travel costs or interference with commitments to work and family (Fig. 1).

Figure 1.

Study design and results of patients following the different therapies after enrolment in the study. Patients with < 75% improvement in Psoriasis Area and Severity Index were considered unresponsive. The group of nonbiological systemic treatments encompasses oral ciclosporin A, oral acitretin, oral methotrexate, and oral psoralen with ultraviolet A irradiation. Grey boxes indicate the total number of screened patients (= 368) and the patients (= 46) who were not enrolled in the study because of contraindications (= 11) or because they did not agree to undergo a phototherapy cycle (= 35). NB-UVB, narrowband ultraviolet B; ETN, etanercept.

Therefore, we enrolled 322 patients with psoriasis (172 women and 150 men, mean age 40·4 years, range 18–84 years), who were treated with NB-UVB phototherapy as the first-line treatment approach (Fig. 1). Of these, 262 (81·4%) achieved PASI 75 and 44 (13·7%) did not. Sixteen patients (5·0%) dropped out from the study for personal reasons.

Twenty-four of the unresponsive patients (7·5%) were switched to conventional systemic therapies for psoriasis, namely ciclosporin A (= 8), methotrexate (= 3), etretinate (= 1) and PUVA therapy (= 12). Only the one patient treated with etretinate did not achieve PASI 75 and he was successfully treated with ciclosporin subsequently.

The remaining 20 patients (6·2%) could not be treated with these agents because of contraindications, previous toxicity or previous inefficiency,[1] and thus received ETN.

Twelve of these 20 patients (60%) responded successfully, whereas the remaining eight (40%) were nonresponsive or poorly responsive (< 75% PASI improvement) after 12 weeks of therapy. These eight were five men and three women with a mean age of 44·9 years (age range 27–63 years) (Table 1). Their skin types were type 1 (= 1), type 2 (= 2), type 3 (= 4) and type 4 (= 1). The mean ± SD PASI score of these patients was 28·3 ± 9·4 at baseline and 21·7 ± 4·9 after the first NB-UVB treatment cycle [P-value not significant (NS)]. Before discontinuation, patients had received 14·9 ± 5·1 J cm−2 phototherapy fractionated over 22·5 ± 1·2 exposures. The mean PASI value was 16·2 ± 9·2 after the 12-week ETN treatment cycle (P-value NS in comparison with PASI after NB-UVB phototherapy) in which each patient received a total dose of 1200 mg of ETN.

Table 1. Clinical details of the eight patients who underwent treatment with narrowband ultraviolet B (NB-UVB) phototherapy, etanercept (ETN) and the combined treatment. Student's t-test was used to compare the Psoriasis Area and Severity Index (PASI) scores of two subsequent treatments
PatientSex/age (years)Skin typeBaselineWeek 12 (after NB-UVB)Week 24 (after ETN)Final (after ETN + NB-UVB)Time to relapse (months)
  1. F, female; M, male; NS, not significant.

Mean44·9 28·321·716·22·42·8
SD12·2 9·44·99·22·81·7
P-value   NSNS< 0·05 

The mean final PASI score after the combined treatment with ETN plus NB-UVB was 2·4 ± 2·8. All patients achieved PASI 75 and three of them cleared completely (Table 1). The patients received 14·6 ± 3·3 exposures resulting in a cumulative dose of 8·4 ± 4·2 J cm−2. The adherence to the combined NB-UVB schedule was very good, with < 10% of missed exposures for each patient. During the combined treatment, patients received a total dose of 450·0 ± 53·5 mg of ETN. The combined treatment was always well tolerated without acute adverse effects. The PSI scores of the irradiated psoriatic plaques were 6·3 ± 2·3 and 0·5 ± 0·8 (< 0·05) before and after the combined treatment, respectively, and the PSI scores of the nonirradiated control lesions were 6·4 ± 2·3 and 5·8 ± 2·5 (P-value NS), respectively (Table 2).

Table 2. Variation of the Psoriasis Severity Index (PSI) score of irradiated and nonirradiated psoriatic plaques of patients treated with narrowband ultraviolet B (NB-UVB) phototherapy and etanercept (ETN)
 UV-exposed plaque (PSI)Nonirradiated control plaque (PSI)
PatientAfter ETNAfter ETN and NB-UVBAfter ETNAfter ETN and NB-UVB
  1. NS, not significant.

P-value < 0·05 NS

In all eight of these patients psoriasis relapsed (PASI > 10) within 2·8 ± 1·7 months (Table 1).


The present clinical investigation demonstrates that a combination of NB-UVB phototherapy and ETN has a synergic effect for clearing the psoriasis of patients who were previously unresponsive to each of the two therapies when administered separately. In addition, we have demonstrated that the clearance of psoriasis was achieved very quickly, with mean cumulative ETN and NB-UVB dosages that were much lower than the respective dosages used when the two treatments were administered individually without success. Nonirradiated control lesions did not improve, which suggests that the response during the combined regimen was not due to a delayed effect of ETN. Unfortunately, the duration of remission at follow-up was rather short in all eight patients. This underscores the fact that these patients had severe and difficult-to-treat psoriasis that was resistant to both therapies alone.

Previous studies have reported that combination of ETN with NB-UVB improves the therapeutic result in untreated patients,[5] and in those patients who were not satisfactorily responsive (at least PASI 75[2] or PASI 90[4]) with ETN alone administered at dosages of 50 mg once weekly[3, 5] or twice weekly.[2] In addition, NB-UVB shortens the clearance phase with ETN 25 mg twice weekly.[3, 5] However, unlike the present investigation, the experimental design of all these earlier studies could not rule out the possibility that similar (or even better) clinical results could have been obtained with NB-UVB phototherapy alone.

In the present and previous investigations,[2-5, 12] combined NB-UVB and ETN was well tolerated and short-term adverse effects were not seen.

The biological basis of the strong synergistic therapeutic effect of NB-UVB with ETN remains to be clarified. Review of experimental data suggests a number of possible mechanisms as follows. NB-UVB irradiation of cell cultures and normal skin induces a dose-dependent increase of TNF-α transcription and release in keratinocytes, T lymphocytes, macrophages, mast cells and dermal fibroblasts.[13, 14]

The increase of TNF-α has a stimulatory effect on keratinocyte proliferation and an overall proinflammatory activity. Therefore it could potentially counteract the anti-inflammatory and antiproliferative effects of NB-UVB, which are thought to play a central role in treating psoriatic skin.[10] Consequently, the selective blockage of TNF-α by ETN could enhance the therapeutic activity of phototherapy. In turn, the whole of the biological effects of phototherapy could potentiate the therapeutic activity of ETN alone by adding to its anti-inflammatory effect.

This hypothesis has been confirmed by an immunohistochemical study[3] that demonstrated a significant reduction of CD1a+, CD4+  and CD8+  cells in the epidermis of patients treated with ETN plus NB-UVB for 6 weeks, in comparison with ETN alone where there was no change of the same cell populations in the dermis, into which NB-UVB does not penetrate.[3] Interestingly, added NB-UVB did not modify the differentiation and proliferation status of psoriatic keratinocytes during ETN therapy, as demonstrated by unaltered immunoreactivity of the differentiation marker involucrin and proliferation-associated Ki-67 expression.[3]

Understandably, there is concern that anti-TNF-α treatments could be hazardous for the UV-exposed skin by contributing to skin carcinogenesis through inducing mutagenic lesions in the epidermis and by reducing antitumour immunity.[15] Indeed, ETN may affect the UVB-induced apoptotic pathways. Low-dose UVB was found to interact with TNF-α to activate TNF receptor-1 (TNF-R1) and induce apoptosis of photodamaged keratinocytes in vitro and in vivo.[13, 16]

In skin treated with erythemogenic doses [2 minimal erythemal doses (MEDs)] of UVB, immunoreactivity of cyclin D1 and p53 was significantly decreased at 24 h, and survivin expression was significantly higher if ETN was administered simultaneously.[17] Furthermore, another anti-TNF-α inhibitor, the chimeric monoclonal antibody infliximab, was found to impair the repair of photoinduced cyclobutane pyrimidine dimer in the spontaneously immortalized, nontumorigenic human keratinocyte cell line HaCaT.[15]

Other experimental findings appear to greatly reduce the concern of possible cocarcinogenicity. Indeed, anti-TNF-α-neutralizing antibodies were not found to modify UV-induced apoptosis, because this is also mediated via ligand-independent activation of TNF-R1,[18] which binds to Fas,[19] and via activation of TNF-R1 by autocrine TNF-α produced by UV-induced activation of c-Jun N-terminal protein kinase.[13, 20] These experimental findings may explain why the expression of p53 in sunburn cells was not modified in the skin of patients with rheumatoid arthritis that was irradiated with 2 MEDs of UVB before and after treatment with the human anti-TNF-α antibody adalimumab.[21]

Epidemiological studies have not definitively assessed the potential carcinogenic risk of the combined treatment in relation to monotherapy with either ETN or NB-UVB.

In a study of 618 patients with psoriasis initially recruited in a randomized controlled trial of ETN lasting 12 weeks, 464 were monitored for up to 84 weeks. The study reported 23 malignancies, of which 14 were nonmelanoma skin cancers (NMSCs) [standardized incidence ratio for any malignancies 1·89, 95% confidence interval (CI) 0·86–3·58],[22] which are most often caused by UV exposure.

Although the observed incidence of squamous cell carcinoma (SCC) in this study was higher than that expected for the general population of the Minnesota-based registry, it was not different from the Arizona-based data.[22] In addition it was not different from the expected number seen in the psoriatic population that at baseline is at an increased risk for SCC, and the degree of risk is positively correlated with the severity of psoriasis.[23] This is likely due to previous excessive exposure to sunlight, PUVA therapy and other immunosuppressive agents.

Patients with rheumatoid arthritis usually receive long-term treatment with TNF-α inhibitors and other immunosuppressive therapies. In one study[24] ETN use in this patient group for as long as 5 years was not found to be associated with an increased incidence of cutaneous SCC. However, a recent meta-analysis[25] of four prospective observational studies[26-29] provided evidence that long-term treatment with TNF-α inhibitors results in a significantly increased risk of developing NMSC (relative risk 1·45, 95% CI 1·15–1·76). In addition, the pooled estimate of two studies[27, 28] for the risk of skin melanoma was 1·79 (95% CI 0·92–2·67).

The discrepancy in the reported risk between patients with psoriasis and patients with rheumatoid arthritis is unknown. One possible explanation is that, unlike most patients with psoriasis, patients with rheumatoid arthritis are treated continuously for long periods with multidrug immunosuppressive regimens. In addition we suggest that dermatologists carefully screen the skin of patients with psoriasis before they initiate treatment with biologics; thus they have the chance to remove skin tumours and premalignant skin lesions.

In light of all of the available clinical and experimental evidence indicating that the combination of treatments may be highly effective but may carry a potential risk of higher carcinogenicity, the question is, if and when the combination treatment should be suggested. The 2006 guidelines for psoriatic arthritis and psoriasis from the National Institute for Health and Clinical and Excellence of England and Wales[30] and the EADF/EADV guidelines for psoriasis of 2009[1] both caution clinicians not to combine anti-TNF and UV therapies.

In the present and previous papers,[2-5, 12] the combination improved the efficacy of ETN alone and was very well tolerated. Therefore, the clinical advantage of the combination regimen could be the higher clearance rate of patients treated with ETN and the decrease of the total time span and cumulative dosages of ETN. This should reduce the risk of drug toxicity and could counterbalance the hazard of increased skin carcinogenesis, if any. In addition, it could strongly reduce the cost of ETN treatment. However, in the present investigation we demonstrated that a combined treatment with NB-UVB and ETN was needed only by a very small portion (2·5%) of the 322 enrolled patients who were not responsive to monotherapies with NB-UVB and ETN. In addition, the same patients presumably could be treated successfully with other biologics or bath-PUVA, modalities that were not included in our study. Therefore the clinical usefulness of the combination seems restricted to very select cases.

Another suggested clinical use[3, 5, 9] is to combine NB-UVB phototherapy over short periods of time to obtain a quicker response to ETN and, possibly, also to decrease the duration of ETN therapy. This approach should be considered for further randomized clinical studies.

In conclusion, ETN has a synergistic activity with NB-UVB phototherapy and their combination is more effective than each therapy alone in the treatment of moderate-to-severe plaque psoriasis, and is well tolerated. Further studies are required to assess the long-term efficacy and safety and to clarify the clinical usefulness in practice.