And as usual, /. poster is stuck on arbitrary methodological constructs s/he learned by rote rather than bothering to understand a) the problem, b) the investigative process in use, or c) the assessable probability of a false outcome without a double-blind administered control group.
And now the answers to a), b), and c) above:
a) basal cell carcinoma varies from almost invisible to looking like acne rosacea. As it gets worse, it usually gets closer to the the acne/rash appearance. It's not generally fatal (although it often occurs on the face and head and can easily metastasize into nearby muscle, lymphatic, or bone tissue with bad outcomes). Traditional treatment is, however often disfiguring because of the necessity of removing all of the skin tissue to get down to the basal layer where the cancer is dividing. Patient ends up with a hole in skin down to the subcutaneous fat or muscle, and often needs a graft. Facial skin grafts are hard to match and can heal badly. What basal cell isn't is variably distributed within the afflicted population. You either have it or you don't, it's very easy and reliable to detect once suspected, and there's really no relevant difference between patients except number and size of carcinomas.
There's also no reason to suspect that a particular treatment will vary in effectiveness over history. IE: if you have ever put a band aid with skin cream or acne cream on ten (or more) randomly-selected BCC patients for a month and recorded the results, you now have a control group that can be compared against any future treatment trail pool of similar size and application method.
Conclusion: no need for a control group in this pilot study - we already know what a band aid will do for BCC.
b) This is a ten-patient pilot study. You do this mostly to see if the proposed treatment is safe for further study or not. If nobody gets sicker then you proceed with a second study. If you also develop some more data (like the 80% cure rate in the ten-sample group here), then great - you have something to put in the funding request for the next study. Trials where 80% are nauseous and develop lesions in seven days probably won't get a follow up. Publishing results at such an early stage of investigation is both a push for funding and for peer review follow-on studies. If such a study makes it into mainstream press (or /. for that matter) then it probably will be misinterpreted by people like the parent poster.
c) given b) above, the investigators are really just looking for a Boolean output signal: either the treatment warrants further study or it doesn't. They got their answer - nobody got sicker and 80% reported improvement. Off to the funding races for a bigger study with control groups and patient outcome tracking.
It's all about risk/reward. Why risk 10 people getting metastasis of their BCC just so that you can control a pilot study that's really more about safety and methodology than efficacy anyway?
It's not about blind, dogmatic adherence to what you learned in middle school about the scientific method.
Speaking as a former lab scientist, this is a perfectly OK methodology for a pilot study.
Speaking as the SO of a basal cell patient, I hope this works so that we can get away from cutting on people with knives to solve the problem. I'd be very curious to hear more about how big the carcinomas were and how well the surrounding healthy tissue filled in the void left by the dying irradiated carcinoma. If nothing else, it should be a lot cheaper treatment than Mohs surgery, and BCC doesn't discriminate based on patient healthcare coverage.