Tuesday, May 29, 2007
I'm traveling the end of this week as well so this will likely be another sparse posting week for me. I'll be at the CSPCP's Advanced Learning in Palliative Medicine conference ( .pdf here) - say Hi if you run into me.
A couple of articles from Annals of Oncology...
First is a look at using the B12/CRP index for prognosticating in advanced cancer. The index ("BCI") is the product of multiplying the serum vitamin B12 level (pmol/l) by the serum CRP level (mg/l) which has been looked at in small, exploratory studies. This article examines the prognostic significance of the BCI in a larger study. It was a prospective, multicenter, UK study looking at 329 adult patients with advanced cancer, no longer undergoing antineoplastic therapy, who were being seen by a palliative care service. As you'd expect, this was a sick bunch of patients: mean age 69 years, most with ECOG performance statuses of 3 or 4, median survival of 42 days. Based on the prior research they divided the subjects into 3 groups: low (better prognosis) BCI (<10,000), intermediate (10-40,000), and high (>40,000). There was a statistically significant trend for survival across the groups - the lowest BCI patients had a median survival of 71 days and the highest had a median survival of 29 days. 90-day mortality was 80% in the highest BCI group and 60% in the lowest.
While this is physiologically interesting, and I'm sure valid (i.e. the higher your BCI the worse your prognosis), I'm not sure how clinically useful this currently is. Is there a big need for an index (particularly one which involves phlebotomy) to differentiate between advanced cancer patients likely to die in one month versus two? This study also embodies one of the toughest aspects of interpreting (even superbly done) research conducted outside of the U.S. - who really are these 'palliative care' patients? There are sufficient differences between U.S. and UK practices and "palliative care"-labeled populations to not know how these data apply to patients that I see, or to US hospice patients with cancer, etc. In addition it would be interesting to see the BCI studied in other populations, particularly ones with longer prognoses (advanced cancer patients undergoing therapy, at the time of diagnosis of metastases, at the time of initiation of "second-line" chemotherapy, etc.).
Christian - as our resident go-to guy on prognostication - any thoughts?
Second is a look at the natural history of malignant ascites . It is a retrospective, single-institution (in the UK) look at characteristics and prognosis of 209 patients with malignant ascites. Mean age was 67 years and at this institution most had their ascites diagnosed concurrently or within the same month as their cancer. Median survival across the entire series was 5.7 months but this varied greatly by cancer type (nearly 2 years for ovarian, a little over 6 months for breast, ~3 months for GI tumors or cancers of unknown primary). Lower albumin and liver metastases both worsened prognosis significantly, although the paper didn't quantify by how much. For me the most interesting aspect of the study was the survival in breast cancer patients: I'd always been taught malignant ascites generally portends a very poor prognosis (ovarian CA aside) - a few months - just like this study found. Conversely one always has the sense that patients with breast cancer do much better than those with most other solid tumors and this study bears this out. Due to this being a single institution study however I wouldn't generalize much about the actual number of months of survival.