Friday, July 17, 2009

Endometrial Cancer: Vaginal Brachytherapy vs Whole Pelvis

In this weeks JCO, the first results of the PORTEC 2 trial are published.

In this randomized trial, 427 patients >= 60 years of age with IC grade 1 or 2, IB grade 3, or any age and IIA disease (excluding grade 3 with >50% myometrial invasion, were randomized between vaginal brachytherapy (7Gy q week x 3, or 30Gy LDR) and WPRT (46Gy, 2Gy/day). Note the inclusion criteria; this excludes the IC grade 3 risk group which at the time of the trial design were at too high of a risk not to treat with WPRT.

This is a QOL report, and not surprisingly, VBT comes out on top, particularly with GI toxicities, which let to better social functioning as well.

This is gratifying, as it proves that less is more as far as the adverse events are concerned. Of course, more importantly what is the efficacy data.

This was presented at ASCO 2008. At 3-years vaginal relapse occurred 0.9% in the VBT arm and 2.0% after EBRT (p=0.97), 3-year pelvic relapse was 3.6% and 0.7% (p=0.03), respectively. 3 year distant relapse was 6.4% in the VBT group and 6.0% in the EBRT group (NS). No difference in 3-year OS (90.4% vs. 90.8% p=0.55) and RFS (89.5% vs. 89.1% p=0.38).

So now it's time to step back to the primary study endpoint, which was vaginal relapse, and unfortunately the stats section is a little vague here, probably because this is a QOL report not the clinical results. What it clear is that this was not designed as a non-inferiority trial for pelvic relapse, which to be honest, is what it probably should have been. If you are trying to say that vaginal brachytherapy is safe and as effective as WPRT, then the endpoint should be either pelvic relapse, distant relapse or overall survival. The pelvic relapses were actually significantly worse with VBT, thought the absolute difference was small (~3%). What is important about this is that these patients are essentially unsalvageable, which may eventually show in the OS numbers (though I wouldn't hold my breath due to the power of the study and competing causes of death in this patient population).

That stated, clearly VBT is an option for these patients, and this study, with relatively minor flaws in comparison to others supports the approach.

Thursday, July 9, 2009

Bevacizumab for NF2 associated AN

An interesting article yesterday at NEJM, in which investigators at MGH gave bevacizumab to 10 patients with NF2 associated acoustic neuromas (vestibular schwannomas). 9/10 had reduction in tumor size after administration and 4/7 evaluable for changes in hearing had an improvement. Very interesting data, and may be useful in large tumors that would be difficult to treat with either surgery or RT.


In this weeks JCO there is an interesting article from investigators in Sweden and Norway reporting a phase II trial of 15Gy x 3 for T1 and T2, medically inoperable NSCLC. Of import, the dose was prescribed to the 67% IDL, meaning that the dose to the center of many of these tumors was much higher.

57 patients were enrolled, the majority were T1 (70%). 3 year OS was 60%, but perhaps more importantly, given their comorbidities, the CSS was 88%. 3 year LC was 92%. 16 patients experienced grade 3 toxicity (28%), and there was one patient with grade 4 dyspnea. No grade 5 toxicities were seen.

This lends even greater support to the practice, and is a great improvement over the prior standard of 70Gy to a postage stamp field to those that are not healthy enough for lobectomy. The next step of course is to look at this in the medically operable patient... stay tuned to ongoing clinical trials on this subject.

Thursday, July 2, 2009


The NEJM this week has in interesting randomized trial on the utility of PET-CT in the preoperative workup of NSCLC.

Investigators from Denmark randomized 189 patients to receive either PET-CT or conventional CT as preoperative staging for NSCLC. Mediastinoscopy was performed in most patients in both arms(94%). After PET–CT, 38/98 patients were classified as having inoperable NSCLC compared to 18/91 patients in the CT arm. 60/98 in the PET–CT arm and 73/91 in the CT arm underwent thoracotomy (p=0.004). 21/60 surgeries in the PET–CT arm and 38/73 in the CT arm were futile (i.e. posiitive mediastinum, distant disease,p=0.05). Survival and number of curative surgeries were no different between the arms.

While it has been long known that PET added significantly to the sensitivity and specificity of lung cancer staging, this is the first study to demonstrate that it results in clinically meaningful results. A reduction in the number of unnecessary thoracotomies would represent a reduction in the cost of treating NSCLC, both financially and in terms of the morbidity and mortality of the procedure itself.