Skip to main content
PMC home page
The BMJ logoLink to The BMJ
. 2000 Jun 24;320(7251):1731. doi: 10.1136/bmj.320.7251.1731

Burns after photodynamic therapy

Drug point gives misleading impression of incidence of burns with temoporfin (Foscan)

Richard Bryce 1
PMCID: PMC1127495  PMID: 10917711

Editor—Hettiaratchy et al report partial thickness skin burns in six out of 14 healthy volunteers injected with temoporfin (Foscan) as part of a phase I pharmacokinetic study carried out on behalf of Scotia Pharmaceuticals at a contract clinical research unit.1 However, their report creates a misleading impression of the true incidence, severity, and overall risk of burns and other photosensitivity reactions with this drug and has led to inaccurate media comment and speculation.

The total number of subjects exposed to Foscan is 957 healthy volunteers and patients, many of whom have been treated with Foscan photodynamic therapy on two or more occasions in clinical studies for a range of different indications. In total, 22 serious adverse drug reactions attributable to photosensitivity including burns have been reported (2.3% of all subjects); this includes the six burns described in the drug point. In clinical studies 15 serious adverse drug reactions involving burns or photosensitivity reactions have been reported in 931 patients injected with Foscan (1.6%). This rate is about half the commonly reported rate of drug related mortality (3-7%) in patients treated with chemotherapy for advanced head and neck cancers.2–6

The table shows the much greater incidence of severe photosensitivity reactions in the volunteers participating in the study referred to by Hettiaratchy et al than in patients with cancer. Furthermore, all 14 of these volunteers reported localised photosensitivity reactions in their infusion arm. The separate pharmacokinetic study in the 23 patients with head and neck cancer, which had an identical study design, reported no serious adverse drug reactions, no burns or localised photosensitivity reactions, and only three generalised photosensitivity reactions of mild or moderate severity (13%); this is similar to the overall rate observed in clinical trials. The pharmacokinetic profile of Foscan in healthy subjects and cancer patients is similar (unpublished data on file).

We conclude that the unusually high incidence and severity of the reactions reported by Hettiaratchy et al are most likely to have been caused by the method of drug administration resulting in photosensitiser being extravasated at the time of infusion. It is well documented that this can lead to delayed and prolonged photosensitivity reactions in the affected tissues. A contributing factor to the severity of the reactions may also have been the extent of volunteer compliance with instructions to avoid prolonged exposure to direct sunlight on discharge from the phase I unit.

In addition, all volunteers participating in the study signed consent forms conforming to the guidelines of the Association of the British Pharmaceutical Industry. No “disclaimers” were sought or obtained. [See correction, 17 June, p 1641.]

Table.

Incidence of burns associated with temoporfin (Foscan)

Phase I pharmokinetic study
 Current Foscan safety database
Volunteers (n=14)1 Patients with cancer (n=23) All patients (n=931) All subjects (n=957)
No of burns 6 0 15 22
Incidence 43% 1.6% 2.3%
Open in a new tab

References

  • 1.Hettiaratchy S, Clarke J, Taubel J, Besa C. Burns after photodynamic therapy. BMJ. 2000;320:1245. doi: 10.1136/bmj.320.7244.1245. . (6 May.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Forastiere AA, Metch B, Schuller DE, Ensley JF, Hutchins LF, Triozzi P, et al. Randomized comparison of cisplatin plus fluorouracil and carboplatin plus flourouracil versus methotrexate in advanced squamous cell carcinoma of the head and neck: a Southwest Oncology study group. J Clin Oncol. 1992;10:1245–1251. doi: 10.1200/JCO.1992.10.8.1245. [DOI] [PubMed] [Google Scholar]
  • 3.Jacobs C, Lyman G, Velez-Garcia E, Sridhar KS, Knight W, Hochster H, et al. A phase III randomized study comparing cisplatin and fluorouracil as single agents and in combination for advanced squamous cell carcinoma of the head and neck. J Clin Oncol. 1992;10:257–263. doi: 10.1200/JCO.1992.10.2.257. [DOI] [PubMed] [Google Scholar]
  • 4.Clavel M, Vermorken JB, Cognetti F, Cappelaere P, de Mulder PH, Schornagel JH, et al. Randomized comparison of cisplatin, methotrexate, bleomycin and vincristine (CABO) versus cisplatin and 5-fluorouracil (CF) versus cisplatin (C) in recurrent or metastatic squamous cell carcinoma of the head and neck. Ann Oncol. 1994;5:521–526. doi: 10.1093/oxfordjournals.annonc.a058906. [DOI] [PubMed] [Google Scholar]
  • 5.Williams SD, Velez-Garcia E, Essessee I, Ratkin G, Birch R, Einhorn LH. Chemotherapy for head and neck cancer. Comparison of cisplatin + vinblastine + bleomycin versus methotrexate. Cancer. 1986;57:18–23. doi: 10.1002/1097-0142(19860101)57:1<18::aid-cncr2820570106>3.0.co;2-v. [DOI] [PubMed] [Google Scholar]
  • 6.Vogl SE, Schoenfeld DA, Kaplan BH, Lerner HJ, Engstrom PF, Horton J. A randomized prospective comparison of methotrexate with a combination of methotrexate, bleomycin and cisplatin in head and neck cancer. Cancer. 1985;56:432–442. doi: 10.1002/1097-0142(19850801)56:3<432::aid-cncr2820560304>3.0.co;2-x. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 Jun 24;320(7251):1731.

Authors' reply (Hettiaratchy and Clarke)

Shehan Hettiaratchy 1,2, John Clarke 1,2

Editor—We were not suggesting that the true incidence of burns with temoporfin (Foscan) is 43%—this was just the incidence in the group of patients referred to us. As Bryce reports, the correct figure is probably around 2%. What we were trying to highlight was that photodynamic agents can cause serious burns and that these may be more severe than conventional burns. This is an unusual occurrence and we state that, to our knowledge, this is the first group of patients given photodynamic therapy ever reported to have burns after exposure to environmental light.

We agree that the incidence of complications in this group was particularly high. This may well have been due to a problem during the administration of the drug. We discussed this directly with Bryce at the time, during Scotia's initial collaboration on the paper. However, this was not a pure extravasation injury. The burns occurred only on exposure to sunlight and photoactivation, two weeks after the drug was given. In addition, other body areas away from the infusion site were also affected, though less severely. This implies that the patients were generally photosensitive because of Foscan. We suspect that there was some leaking out of the drug on administration, leading to high tissue concentrations around the infusion site. On photoactivation the most severe reaction occurred here, resulting in the worst burns.

Irrespective of whether there was some problem with the administration of the drug, these patients had serious burns and we believe that the causative agent was Foscan. This is a rare but serious complication that the medical community needs to be aware of. It is important for clinicians using photosensitisers to know that complications can occur and for burns surgeons to know that burns induced by photodynamic therapy may behave differently from conventional thermal injuries.

Footnotes

Competing interests: Mr Clarke has provided medical reports on the injuries and the initial management of the six patients reported on in the drug point. No fee was charged.

BMJ. 2000 Jun 24;320(7251):1731.

Authors' reply (Täubel and Besa)

Jörg Täubel 1, Chola Besa 1

Editor—We have until now been cautious to enter into the debate which has arisen concerning the article of 6 May.2-1 In view, however, of the legitimate interest doctors and the public have in the issues raised, and in view of the criticism of ourselves and Charterhouse Clinical Research Unit which has arisen as a result of the contents and comments on the BMJ article, it is important that we now set out the facts and answer some of the criticisms that have been made.

Notwithstanding that we were named as coauthors, the final form of the article published in the BMJ had not been approved or seen by us. Amendments which we had made were apparently rejected at some stage and an addition was made which was simply incorrect—namely, that the volunteers had signed disclaimers and were therefore not entitled to compensation. In fact, no fault compensation insurance was put in place by both ourselves and the drug company in accordance with our standard practice.

The formulation of Foscan (temoporfin) used in the trial reported in the BMJ article of 6 May was, we were given to understand by the drug company, different from that used in previous trials. Specifically, a new solvent was added to the drug so that it would be more soluble and less painful to administer. This factor is relevant to an evaluation of the results of the trial and any comparison of the results with those in previous trials.

While seven of the volunteers were treated by Mr Clarke at Chelsea and Westminster Hospital Burns Unit, they were the only volunteers it was appropriate to refer for specialist treatment. In fact, of the 14 volunteers on whom the new formulation of the drug was tested, all 14 suffered some form of burn on the forearm into which the drug was infused. In 13 cases the burn tracked the vein into which the drug was injected and was long and oval in shape, starting close to the site of infusion and extending along the forearm (although in one instance this was obscured by the extent of the burns to the forearm). In the other case, although there was a burn which appeared to follow the infusion vein, the burn only started to occur towards the elbow, away from the site of injection.

Charterhouse was especially careful to ensure that the drug was correctly injected into the vein of each volunteer. None of the infusions was given perivenally, and no drug was spilt on the skin as suggested in one press article. Charterhouse is very experienced at carrying out infusions which demand that there is no accidental injection into surrounding tissues and performs them on a regular basis. We applied an appropriate standard technique of administering the drug into a vein of the forearm. We did not attempt to inject into veins in the antecubital fossa as they are best avoided because of the risk of interarterial injection (aberant ulnar artery) and the possibility of median nerve damage, particularly as the drug had the potential to cause local damage. Numerous safeguards in our routine procedures prevent us from accidentally injecting any drug into surrounding tissue. These include (a) injecting into superficial veins in the forearm to provide a visual control as to whether or not the canula is correctly placed; (b) the use of pumps with an inbuilt pressure sensor that immediately stops the pump if there is an increase in pressure, which would occur if a canula slipped out of a vein and into the surrounding tissue (which would cause increased resistance); (c) the flushing of canulas with water for injection before starting the infusion to provide a visual and sensual check as to whether there is any resistance as a result of either a blood clot forming at the tip of the canula or the tip being displaced into tissue; and (d) personal supervision of each infusion by staff with the requisite expertise and experience.

Each safeguard detailed above was followed in respect of every injection given in the Foscan trial reported in the BMJ. Each infusion was personally supervised by one of us. In each and every case, there was no evidence or indication that the drug was administered perivenally to any of the 14 volunteers. If a perivenal injection had been given, one would expect the resultant burn to be approximately circular in shape around the site of infusion and not a long oval burn tracking the line of the vein.

We consider that the most probable cause of the burn to the infusion forearm suffered by each of the volunteers was that the drug, having being injected into the vein, then leaked from the vein by an unknown mechanism (a hypothesis supported by the number of volunteers on whom the burn was observed). We do not know whether there is any connection between the leakage and the new solvent used in the trial we performed, but so far as we are aware the characteristic shape of the burns we recorded were not recorded in previous trials of the drug.

It is unfortunate that the BMJ did not send us a copy of its standard form on competing interests since we would not have hesitated to provide the information requested. That said, we accept that it would nevertheless have been appropriate for us to ensure that the nature of our involvement in the relevant trial was made explicit, and the version of the article which we had approved did not do so. We are employed by Charterhouse Clinical Research Unit Limited, which conducted the drug trial on behalf of the relevant drug company, and personally supervised the trial. Charterhouse entered into a contract with the sponsoring drug company relating to the performance of the Foscan trial and received remuneration for conducting the study.

We are convinced that Foscan is a very promising drug candidate which will be valuable for the treatment of certain cancers. We have simply carried out a trial which, as is often the case, raised certain queries which may require further investigation. The primary purpose of carrying out a drug trial is to advance medical knowledge about the side effects of a particular drug formulation where this is not already known. As doctors we would not wish the continued development of Foscan to be in any way impaired by the BMJ article of 6 May and the debate which followed.

References

  • 2-1.Hettiaratchy S, Clarke J, Taubel J, Besa C. Burns after photodynamic therapy. BMJ. 2000;320:1245. doi: 10.1136/bmj.320.7244.1245. . (6 May.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Jun 24;320(7251):1731.

Editor's reply

Richard Smith

Editor—We, the BMJ, did not do as well as we would have liked in publishing this drug point. I must make clear, however, that we did send a proof to the corresponding author and did not make any changes to the corrected proof.

Our policy is to ask authors reporting an adverse event associated with a drug to contact the manufacturers and ask for data on whether they have had reports on similar events. We did that, but we did not take the next step of including the data in the paper. That was a mistake, and we have taken steps to avoid it happening again.

We should have sent the authors a copy of our standard form on competing interests and published their answers. Mr Clarke and Drs Täubel and Besa have now declared their competing interests. Although many journals ask authors to declare competing interests, our own experience is that virtually nobody volunteers a competing interest unless presented with a set of explicit questions. That is why we now have a form that includes explicit questions on financial competing interests (available at http://www.bmj.com/advice/5.html), and our experience is that authors do respond to this form. I am sorry that we failed to send the form to the authors on this occasion.

The letter from Drs Täubel and Besa is considerably longer than we normally allow. We apologise to readers for this, but we wanted to publish these letters and explain our part in the story promptly. When it became clear that the authors had not been communicating well among themselves we postponed publication to allow Drs Täubel and Besa to respond separately. Their letter arrived, via their lawyers, at this length at the last possible moment for inclusion in this week's journal, allowing no time for negotiation over editing. We thought that it was better to publish their letter at this length together with the other letters than to publish it separately next week.

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES