Can UVB Phototherapy Cause Cancer?

This report attempts an answer, but it isn’t straightforward.

Can UVB, as administered in a dermatologist’s office, cause cancer? This is a question that’s been examined a few times in the medical literature, and comes up on the psoriasis newsgroup every once-in-a-while. I’ve been involved in at least a few of those discussions (sometimes heated, especially when people who claim there’s no risk whatsoever are involved). After one of them, I exchanged a couple of letters with a prominent psoriasis researcher (who wished to remain nameless). Much of this article is copied from the last letter I sent to him, in May of 2002 (which, unfortunately, went unanswered).

We know that sunlight can cause various forms of skin cancer, and we know that it does so at wavelengths of light which are also emitted by the UVB bulbs (both wideband and narrowband) in the lightbooths used to treat psoriasis. But, since UVB phototherapy in a dermatologist’s office is controlled and limited — in contrast to sunbathing — the question of whether or not UVB phototherapy increases a person’s risk of skin cancer, as sunbathing can, needed to be addressed.

Let me now state that I agree completely that UVB is one of the safest treatments there is for moderate-to-severe psoriasis, but “safest” is a relative term. It is, most assuredly, less risky than several therapies available, but even being the safest does not mean that there’s no risk whatsoever. And, of course, there are people who cannot use UVB, or who get few or no results from it, and so must turn to riskier therapies to control the disease.

When writing that last letter, I did a bunch of digging through Medline, but the results of my digging were, to say the least, disappointing. The doctor had written to me of “numerous” studies examining UVB phototherapy and showing a lack of carcinogenic potential, and repetition and confirmation of Pittelkow’s study (more on this, below), but I was unable to find any of this, except for one study which only went back 13 years, and did not compare melanoma rates[1], and another which didn’t claim much more than “it is unlikely that the psoriasis patients treated [with extensive UVB] in the past have an increased prevalence of premalignant and malignant skin lesions” (emphasis mine).[2]

I used at least two dozen different combinations of keywords while searching (“cancer,” “carcinogenic,” “carcinoma,” “mutagenic,” “UV,” “UVB,” “UV-B,” “ultraviolet,” etc.), with the addition of “psoriasis” to all searches, and came up nearly empty (except for the above) after ignoring all of the results related only to PUVA (of which there were hundreds — PUVA carries a well-known cancer risk).

In fact, as I see it from my searches, the “consensus” appears to be that clinical-setting UVB phototherapy does indeed carry some small level of cancer risk. The greatest risks appear to be for the face[3], and the male genitalia[4,5], and so of course these areas are protected when possible (then again, the fact that they need protection implies a risk of sorts).

And the other articles I found were, for the most part, rather tentative:

· Reducing the number of UVB sessions may mean “less risk” of cancer[6]

· Narrowband UVB has “possibly lower long-term cancer risk” than PUVA[7]

· “Far erythemogenic” narrowband UVB therapy is “possibly” less carcinogenic than near erythemogenic[8]

· Both PUVA and broad-band UVB “carry a risk of carcinogenesis”[9]

· UVB, as well as arsenic, X-rays, tar, and methotrexate, are assumed to be carcinogenic risk factors[10]

· “…maximum safe cumulative doses of UVA or UVB have not yet been established”[11]

· UVB is “probably less carcinogenic” than PUVA[12]

· Narrowband UVB reduces (but doesn’t eliminate) exposure to carcinogenic wavelengths[13]

There was also an article about the use of UVB in treating atopic dermatitis. The researchers found an effective dose was 1 joule/cm2, and said that they calculated a relative risk of non-melanoma skin cancer after 15 years of treatment of 1.15 by age 60.[14] They also stated that the dosage for atopic dermatitis was one-fourth that typically used in the UVB treatment of psoriasis. It seems counterintuitive to me to think that the risks would go down with a higher dosage of UVB.

Another aspect of the question is that of geographic location and duration. In the Netherlands, decades-long “maintenance” UVB therapy was found to create a cancer risk several times that of “outdoor occupations.”[15] In Sweden, therapeutic UVB exposure was found to be on par with outdoor work, and cancer rates were expected to rise to similar levels if UVB therapy was expanded, despite a lack of such findings in a retrospecitve study.[16] And although the abstract for Dr. Pittelkow’s psoriasis study[17] doesn’t mention it, the contemporaneous study on atopic dermatitis limits itself to drawing conclusions only about “selected populations of the United States.”[18]

Dr. Pittelkow’s (et al) psoriasis study, published in 1981, looked into the cancer rates of 280 patients treated at the Mayo Clinic between 1950 and 1954. The patients were treated with the Goeckerman regimen, which consists of coal tar and UVB, and compared to patients who received coal tar alone. After 25 years, they found no appreciable increase in skin cancers after 25 years, and this result is widely quoted as the “proof” that UVB isn’t cancer-causing. But, as noted in Dr. Pittelkow’s atopic study, there are geographic reasons for not generalizing from these results to the population world-wide. And the concurrent use of coal tar adds a confounding factor which should have been addressed.

But perhaps the most “damning” abstract I found was a meta-analysis and review of the literature published up to 1999, which indicated

The available evidence is insufficient for quantifying the [excess incidence] of nonmelanoma skin cancer in patients with psoriasistreated with UV-B radiation. However, it seems unlikely that theexcess risk exceeds 2% per year.[19]

Besides the PUVA studies, another thing I ignored in my searches was the many articles which discussed the carcinogenic properties of UVB on animals. Even narrowband UVB lamps show a dose-response relationship for cancer in hairless mice.[20] But, even though the mechanisms might be the same[21], only humans get psoriasis, so something different must be going on (at the very least, the mice are probably “blasted” with UV, in much higher doses than are used for psoriasis phototherapy).

Now, none of the above explicity states that phototherapeutic UVB, in any form, does, indeed, cause cancer. But a large fraction of abstracts of studies which examine this question do indicate that a cancer risk is associated with UVB, at least in the minds of those authors.

After doing these searches, a friend of mine sent me a copy of the NPF Bulletin article, “Is UVB Administered in Phototherapy Carcinogenic?”[22] My friend also sent a copy of the NPF Forum article with the same title, which is somewhat more speculative, and includes a section on narrow-band UVB which the Bulletin article did not.[23] This section includes:

The new technique of narrow-ban [sic] phototherapy is aimed at maintaining UVB’s therapeutic action while decreasing the known risks associated with short wavelengths of UVB.

Knowing that it’s the shorter wavelengths which are more carcinogenic, and that the most-effective wavelengths for psoriasis appear to be around 312 nm, is important. Judging from the Philips lamp data sheet available on their and other web sites, the wide-band lamps go well below 300 nm wavelength. While the power does drop off sharply, even 280 nm from those lamps has roughly 20% of the peak power, and 290 nm has somewhere around 60% of peak. This appears to me to be a significant amount of light. And, a few sentences later in the more-detailed (NPF Forum) article:

Researchers hope this treatment [narrowband UVB] will “do less damage and give more of the benefits” of UVB, said Dr. Pittelkow. However, “The full story is not in,” he added. Further research must determine whether narrow-band phototherapy, which is administered in greater dosage, is indeed a lower-risk method of psoriasis treatment.

So now we read Dr. Pittelkow himself implying that wide-band phototherapy does, indeed, have a cancer risk associated with it, regardless of what he’d found sixteen years earlier. Perhaps he’s being quoted out of context, and it wasn’t the cancer risks he was talking about, but instead the burning risks? Or risks of premature skin aging? Tough to say, but given the article’s title, and the reputation of the National Psoriasis Foundation, it seems unreasonable to think so.

So, I’ve seen the abstracts for the four studies mentioned specifically in the NPF Forum article, along with some of Dr. Robert Stern’s studies on PUVA.[1,2,3,17,18] But, both the doctor I wrote this letter to and the NPF Forum article mention “several studies” and imply that there are more. I can’t find them, after what was truly intensive searching. But, as I readily admit, my searches are limited to what’s online, and my ability to “follow through” (full text of articles) is even more limited.

I did this searching because I am very interested in getting to the bottom of this, one way or the other. It doesn’t matter to me whether I’m right or if people making claims that UVB is completely safe are right (or neither or both), what matters is getting the correct information out to anyone who asks or comes looking for it.

That said, given the diversity of the results in those studies, I cannot bring myself around to calling the absence of carcinogenicy in UVB phototherapy “definitive” or “conclusive” or anything of the sort. Not at this point in time. Not when a cancer risk appears to be assumed in at least nine other papers since Pittelkow. And not when Dr. Pittelkow himself has apparently made that implication, as well.

And having done all this searching, I think I’ve found a reason why I’m not comfortable with such a conclusion.

I have seen two items which hint at a lower incidence of skin cancers in psoriatics in general. The National Psoriasis Foundation sent out a questionaire, which found that 85% of psoriatics felt a need to hide their psoriasis, and 85% also avoid sports or swimming.[24] And a Swedish study found that the incidence of malignant melanomas in psoriasis patients was one-third that of their general population.[25] It seems to me, then, that comparing post-phototherapy cancer rates to that of the general population, as Dr. Pittlekow did 21 years ago, is not a valid methodology with which to determine the cancinogenic risk of UVB itself in psoriatics. While the subjects in these studies tend to be carefully age-, sex-, or skin-type-matched with controls, they are never, as far as I’ve seen, matched with regards to exposure to sunlight or UVB in general.

To compare apples to apples, what should be tested is the rate of the three major types of skin cancer after various periods of UVB phototherapy (and a long follow-up time), compared to the rate of those cancers in psoriatics matched for average coverage (perhaps somehow based on PASI score), but who have never received phototherapy in a clinical setting. UV exposure from the sun or tanning beds should be random and about equal in both groups. I would guess that those who received clinical UVB phototherapy will have a higher incidence of cancer, and that the control group will have an overall lower rate of cancers than the general population, which consists of people who generally aren’t mortified by the idea of wearing t-shirts and shorts.

If this is true, then finding that the cancer risk after UVB phototherapy is consistent with the general population would mean that there is, indeed, a measurable and probably-significant risk in the therapy itself. And even more so with PUVA. The NPF Forum article, in implying that the known PUVA-cancer risk disconfirms the idea that psoriasis is somehow cancer-protective, ignores the possibility that psoriatics simply don’t get as much exposure as normal folk. It’s not that psoriatic skin is somehow “immune” to cancers, it’s that we hide from a major cancer-causing agent, the Sun.

Now, I’ve also read a couple of abstracts which may be seen as disagreeing with this idea, both of them from Dr. Stern.[26,27] He says in one,

Contrary to long-standing beliefs, our findings support the hypothesis that even after allowing for possible greater exposures to cutaneous carcinogens, the risk of nonmelanoma skin cancer in patients with psoriasis is at least as great as in the general population.

However, I, perosnally, cannot justify ignoring “long-standing beliefs” based on the results of just two studies with the same lead author. And, while not statistically significant, Larko’s finding of just 5.9% of psoriatics with cancerous lesions versus 10.1% of controls seems to me to merit further research.[2] A simpler study than I proposed above, even: Just measure the amount of skin cancer in psoriatics who’ve had symptoms for, say, 40 years, without ever undergoing either PUVA or UVB, and compare the rate to a similarly-aged non-psoriatic control population who’ve never undergone phototherapy. To reiterate, my guess would be that the social stigma of psoriasis would ensure a lower cancer rate, without the need to speculate about a possible biological “carcinoprotective” trait.

Again, I feel it’s important to note that even if a small risk of cancer does exist with UVB phototherapy in its varied forms, it is, indeed, one of the safest treatments available for moderate-to-severe psoriasis. This article is not intended to be a fear-mongering scare tactic to get patients to drop UVB like a hot potato. Far from it, the intent is that people use this very safe therapy while armed with all the information about it that can be found and to point out to some that its safety is not a good reason to downplay any of the possible risks until they are shown, definitively, to not exist. Nothing I’ve seen to date does that for the cancer risk.


Note: In the majority of the below footnotes, I reference the papers for the information contained in their abstracts only. I am aware that the full-text articles may provide valuable additional information, but I am not able to access the full text for most of these as of this writing.

1.The carcinogenic risk of treatments for severe psoriasis. Photochemotherapy Follow-up Study.” Stern and Laird, Cancer 1994 Jun 1;73(11):2759-64

2.Is UVB treatment of psoriasis safe? A study of extensively UVB-treated psoriasis patients compared with a matched control group.” Larkö and Swanbeck, Acta Dermato-Venereologica 1982;62(6):507-12

3.Risk evaluation of UVB therapy for psoriasis: comparison of calculated risk for UVB therapy and observed risk in PUVA-treated patients.” Slaper, et al, Photo-Dermatology 1986 Oct;3(5):271-83

4.PUVA, UVB, psoriasis, and nonmelanoma skin cancer.” Studniberg and Weller, Journal of the American Academy of Dermatology 1993 Dec;29(6):1013-22

5.Genital tumors among men with psoriasis exposed to psoralens and ultraviolet A radiation (PUVA) and ultraviolet B radiation. The Photochemotherapy Follow-up Study.” Stern, The New England Journal of Medicine 1990 Apr 19;322(16):1093-7

6.Pharmacoeconomic evaluation of calcipotriol (Daivonex/Dovonex) and UVB phototherapy in the treatment of psoriasis: a Markov model for The Netherlands.” de Rie, et al., Dermatology (Basel, Switzerland) 2001;202(1):38-43

7.Narrowband UV-B phototherapy vs photochemotherapy in the treatment of chronic plaque-type psoriasis: a paired comparison study.” Tanew, et al, Archives of Dermatology 1999 May;135(5):519-24

8.Comparison of phototherapy with near vs. far erythemogenic doses of narrow-band ultraviolet B in patients with psoriasis.” Hofer, et al, The British Journal of Dermatology 1998 Jan;138(1):96-100

9.Biological effects of narrow-band (311 nm TL01) UVB irradiation: a review.” el-Ghorr and Norval, Journal of Photochemistry and Photobiology 1997 Apr;38(2-3):99-106

10.Skin tumors in photochemotherapy for psoriasis: a single-center follow-up of 496 patients.” Maier, et al, Dermatology (Basel, Switzerland) 1996;193(3):185-91

11.Ten-year experience of phototherapy in Yonsei Medical Center.” Park, et al, Yonsei Medical Journal 1996 Dec;37(6):392-6

12.Comparison of narrow-band UV-B phototherapy and PUVA photochemotherapy in the treatment of psoriasis.” Van Weelden, et al, Acta Dermato-Venereologica 1990;70(3):212-5

13.311 nm UVB phototherapy — an effective treatment for psoriasis.” Green, et al, The British Journal of Dermatology 1988 Dec;119(6):691-6

14.Phototherapy of atopic dermatitis with ultraviolet radiation.” Jekler, Acta Dermato-Venereologica Supplementum 1992;171:1-37

15.UVB doses in maintenance psoriasis phototherapy versus solar UVB exposure.” Schothorst, et al, Photo-Dermatology 1985 Aug;2(4):213-20

16.Phototherapy of psoriasis — clinical aspects and risk evaluation.” Larkö, Acta Dermato-Venereologica Supplementum 1982;103:1-42

17.Skin cancer in patients with psoriasis treated with coal tar. A 25-year follow-up study.” Pittlekow, et al., Archives of Dermatology 1981 Aug;117(8):465-8

18.Incidence of skin cancers in patients with atopic dermatitis treated with coal tar. A 25-year follow-up study.” Maughan, et al, Journal of the American Academy of Dermatology 1980 Dec;3(6):612-5

19.Treatment with UV-B for psoriasis and nonmelanoma skin cancer: a systematic review of the literature.” Pasker-de Jong, et al, Archives of Dermatology 1999 Jul;135(7):834-40

20.The dose-response relationship for tumourigenesis by UV radiation in the region 311-312 nm.” Sterenborg, et al, Journal of Photochemistry and Photobiology 1988 Sep;2(2):179-94

21.Ultraviolet radiation in skin ageing and carcinogenesis: the role of retinoids for treatment and prevention.” Oikarinen, et al, Annals of Medicine 1991;23(5):497-505

22. “Is UVB Administered in Phototherapy Carcinogenic?” Lebwohl and Koo, National Psoriasis Foundation Bulletin 1997 Nov/Dec:7

23. “Is UVB Administered in Phototherapy Carcinogenic?” Lebwohl and Koo, National Psoriasis Foundation Forum 1997 Fall:3,6

24.News and Notices, AAD Dispatches” The National Psoriasis Foundation. Web Site

25.Cancer risk in a population-based cohort of patients hospitalized for psoriasis in Sweden.” Boffetta, et al., The Journal of Investigative Dermatology 2001 Dec;117(6):1531-7

26.Psoriasis and susceptibility to nonmelanoma skin cancer.” Stern, et al, Journal of the American Academy of Dermatology 1985 Jan;12(1 Pt 1):67-73

27.Psoriasis and the risk of cancer.” Stern, et al, The Journal of Investigative Dermatology 1982 Feb;78(2):147-9