2000miler

Marion – I remembered using the Wayback Machine years ago when I was doing genealogy. I tried it this morning for a couple of 2006 crawls and while I can get to the Discussion Boards’ tables of content, clicking on individual topics brings up the message, “The Wayback Machine does not have this URL” and it defaults to the current web pages.

This doesn’t appear to be the case for more recent crawls when the machine apparently went deeper into the web site. For example, I pulled up a 1/15/13 crawl and the latest posts were for 1/15/13.

I’ll have to check different crawl dates to see when they shifted from the limited crawl to a full crawl.

Bruce

The following survival plot compares cc.org patients who have had radioembolization and traditional chemo with patients who have had radioembolization, traditional chemo and targeted agents.

My database only has 11 patients with just chemo and 8 with chemo and targeted agents, so the plots are ragged and definitely not statisticallhy significant, as illustrated by a “p” of 0.916. However, Percy wanted to see this comparison so I am including it in this topic for all to see.

Bruce

Below is the first survival plot that I have made showing the impact of resection. Instead of plotting resection against no resection, I decided to divide no resection into two groups, those where resection was denied a patient because of distant mets and where it was denied for other reasons such as blood vessel involvement. Since distant mets (M1) are responsible for the highest stage level of cc (IVB for intrahepatic and perihilar cc and IV for extrahepatic cc), I assumed there would be a considerable separation in the distant mets vs. no distant mets curves. Surprisingly, there was very little.

Statistics for the data was chisq = 63.5, 2 degrees of freedom, and p=1.64e-14. Medians survivals were 3.373 years for resections, 0.914 years for no resection with no distant mets, and 0.753 years for no resection with distant mets.

Bruce

Hi Percy,

I am producing these plots as I gather sufficient data for the plot to be statistically significant. I have generated other plots and found them to be statistically insignificant (p>.05), so I haven’t posted them.

A this time I am trying to create a plot comparing survival for the following three variables:

(1) Patients who were initially diagnosed with distant mets and could not be resected.
(2) Patients who were initially diagnosed with other than distant mets and could not be resected.
(3) Patients who were resected both initially and at a later time, but did not have a transplant.

Eventually, I will get around to plots which include comparisons between the three types of CC (intrahepatic, perihilar, and extrahepatic) types of chemo, radioembolization, etc., but I need to wait to build up the database before doing this, because I know I just do not have enough data for the more detailed categories.

Of course, if you would like to search the posts and develop an Excel spreadsheet with the variables in which you are interested, I would be more than happy to enter them into the R program and produce the survival curves for you.

Bruce

Thanks Eli, I record CC type, surgical margins and node involvement, but not surgical complications.

About 15% of the members of cc.org do not post about themselves or family members or friends that have pure cc. Some of these members discuss medical issues or people with gallbladder cancer or mixed cancers, etc. About 6% of the members include those with multiple IDs or those who discuss the same patient as other members. Of the remainder, a little under half of the members don’t report the CC type of give me enough clues to figure it out. As a result I can identify CC type for about 43% of the members. I have surgeries for about 26% of the members, and surgical margins for about 70%of the surgeries. I have negative and positive node data, but I am missing a lot of those, although it might be safe to assume that most nodes are negative if they are not mentioned in the posts.

As you suggest, my sample right now probably is not large enough to test your hypotheses, but I’ll check them out and see what I get.

Bruce

Hi Percy,

So far, I have compiled data for 15% of the patients on these discussion boards and have found only 12 who have had radioembolization. Most of the 12 did not use the term “radioembolization,” but called it brachytherapy (3), Sir-spheres (5), Theraspheres (3), and spheres (1). Based on the present 2,438 members on these boards, a rough estimate of the total would be 80.

I might have missed some of the treatments, especially if they were done well after a patient was dx with cc and especially if the member had many posts.

The breakdown of the patients who had radioembolization is as follows:

Brachytherapy:
M, 46, Transplant, still alive at last post, 4.38 years after dx
M, 58, No surgery, lived 5.41 years after dx
F, 66, Transplant, still alive at last post, 1.49 years aftr dx

Sir-spheres:
F, 45, No surgery, lived 1.51 years after dx
F, 58, No surgery, lived 3.21 years after dx
F, 51, No surgery, lived 0.95 years after dx
F, 56, No surgery, lived 2.47 years after dx
M, 67, Resection, lived 7.64 years after dx

Theraspheres:
M, 77, No surgery, lived 0.73 years after dx
F, 25, No surgery, still alive at last post, 1.56 years after dx
F, ?, Transplant, still alive at last post, 1.24 years after dx

Spheres:
M, 60, No surgery, lived 2.31 years after dx.

I don’t have any data on duration between start of treatment & final outcome.

Bruce

Hi Eli,

The R program can do multvariate analysis, but I don’t think it is within it’s survival package. The survival package is designed to combine times to actual deaths and times to when patients leave the study while still alive, referred to as right-censored data. I have done multivariate analysis before using Lotus, but right now, I don’t have a feel on how I would combine actual deaths and right-censored data within a mutivariate analysis.

I was surprised by my results, because I haven’t seen any cc survival studies involving sex and age. I did notice that the Cisplatin plus Gemcitabine versus Gemcitabine for Biliary Tract Cancer Phase III Clinical Trial balanced age and sex in their studies. Median age for Gem was 63.2 yrs. and 63.9 yrs for GemCis. Also, there were 108 women and 98 men for Gem, while there were 108 women and 96 men for GemCis. So it appears they were trying to negate any possible sex and age influence on the results. The survival curves shown in Fig. 2 of their paper were only separated by a few months, which was much closer than the curves I show in the two previous plots.

I have a lot more men than women in my data mostlly because about four times as many wives post about their husbands to these boards than husbands post about their wives. Daughters and sons posts about evenly for their parents, but daughters outnumber sons in the posts by over four to one.

Somewhere in the future, I will show survival curves for more than one variable, so I could include sex, age, and resection status in one survival plot which would allow comparisons. The problem is that as I include more variables, the sample size for each condition decreases and p increases above 0.05, so I lose statistical significance. I correct this by compiling more data from the posts, but this is very time consuming. Also, I spend a lot of time looking online for possible deaths of members who no longer post because the survival analyses starts to lose accuracy if too many right-adjusted data points are used.

Bruce

[4/25/13 – Eli, since I posted this I learned that the Cox Regression Model is used in survival analysis to handle the right-censored data for multivariate analyisis. The R Survival Package includes Cox regression analysis and I will eventually use it to do the analysis you suggested.]