Dear,

On simple note, you need following variables : Survival time, Events 

Parameters required

- cut off date for follow up or censoring date, or study end point date. (A)

-date of remission/ or date of surgery(B)

-date of diagnosis(C)

For DFS, survival time:  A-B

For OS survival time: A-C

Designate your relapsed patients 1 and rest 0,( For DFS) Event  will be 1

Designate your died patients 1 and rest 0, (For OS) Event will be 1.

Be careful to exclude the death other than disease burden.

Subsiquntly use any statistical package to get Kaplan Meier curves followed by log rank test  to compare between the groups

Best Wishes..

Surender

Search for time to event analysis.

A few things:

* Dividing patients into groups based on their outcome does not make sense, because patients can change categories, depending on your follow up. E.g a patient alive and disease free at 5yr post-surgery might have a recurrence at 5 yr and 1 mo.

* I am not sure if you are doing this, but you should also include those with less than 5 years of follow up. So, do include patients who are alive and disease free who had surgery only last month. The good thing about time to event analysis is that every bit of follow up can be included.

* The method to estimate your DFS at 5 years could be the Kaplan-Meier method (google will give many hits). For this, the data should be in the following shape:

3 columns:

ID = patient Id

TimeBetweenSurgeryAndLastFU = number of days between surgery and the last follow up

Status = 0 for patients alive at their last FU, and 1 at the date of death or date of recurrence (whichever came first)

You could calculate 5yr DFS by hand, however there are various software packages are there to calculate the DFS, and provide a Kaplan-Meier plot. Any statistical package you have available at your university will be able to do this.

I use Armitage and Berry, Statistical Methods in Medical Research, 2nd Ed. Blackwell Scientific Publications, Oxford, 1987.  I suggest reviewing Chapter 14 Survival Analysis. Substitute disease recurrence for death.  Plot all data from all patients from entry into the trial to recurrence.  From this plot, you will be able to quickly visualize how many patients are still disease free at 5 years, or any other time of your choosing.  Note: some patients may die before they are diagnosed with a recurrence or they are lost to follow up.  If so, their data is considered censored.

DFS is sometimes plotted. You must define time 0. This might be time from diagnosis or the date of the initial surgery.  If you plot x (number of patients) vs. y (vs. time since surgery) then you will have a curve.  

It is often useful to plot both DFS and OS (overall survival).   Overall survival is useful as sometimes you will see effects of treatment that are not expected. For instance, if the surgery or its complications killed a lot of patients (Hopefully not!) then this would show up. 

DFS=(alive without disease) / all patients

OS=( alive without disease + those with recurrence)/ all patients

Number of patients (X axis) vs. time (Y axis)

            Define time 0 which might be time from surgery or time from diagnosis. 

This might also be interesting: Causes of death:

            Died with disease but from another cause

            Died without evidence of recurrence

            Died from recurrent disease or complication of  treatment of either the original problem or treatment of the recurrence;  

How you decide which patients to include and which to exclude all needs to be stated in your methods.  You probably know this, but you need a demographic table showing all the relevant data such as age/sex/relevant co-morbidities, etc.  Are you including all patients who had surgery? What about those that consulted you and did not have surgery (probably omit)? 

Make sure ALL patients are included.  If you omit someone, make sure you say why such as "lost to follow up",  "consent withdrawn", "cause of death not specified", etc. 

It can be complicated to figure out what the cause of death was and whether this was related to the original disease or not.  For instance, a patient with lung cancer might have pneumonia listed as the cause of death.  There is a National Death Index that can provide the cause of death information for a small fee. 

 Kaplan Meier is the most common analysis but I don't have enough statistical background to advise one over the other.  

It's really worth working with a statistician as otherwise your article might get thrown out for some obvious statistical problem. 

Dear,

On simple note, you need following variables : Survival time, Events 

Parameters required

- cut off date for follow up or censoring date, or study end point date. (A)

-date of remission/ or date of surgery(B)

-date of diagnosis(C)

For DFS, survival time:  A-B

For OS survival time: A-C

Designate your relapsed patients 1 and rest 0,( For DFS) Event  will be 1

Designate your died patients 1 and rest 0, (For OS) Event will be 1.

Be careful to exclude the death other than disease burden.

Subsiquntly use any statistical package to get Kaplan Meier curves followed by log rank test  to compare between the groups

Best Wishes..

Surender

There may be several models, some  of them have been  suggested as above. i  may refer my recent research paper  on diagnosis  n prognosis  of cancer patients by  taking help of  ancient wisdom which is published in Indian Journal of Research  in Ancient Medicine & Yoga, No. 8 vol. 2 April- june, 2015. I t may possibly  be helpful in  dividing groups of cancer patients. More over if it is useful in any way,  there may opening  for searching clues from ancient wisdom in the field of oncology.

The minimum requirement for calculating DFS is to have good documentation on disease recurrence: local or distant or both (considered as events or failures) or none (no event). There may be information on deaths of patients not due to cancer. You have to take a call on this whether to treat these also as "events". These should technically be treated as censored.

Data on all patients whether alive or dead or with or without disease at the end of follow up must be included. A substantial proportion of your patients must have a potential follow up of 5 years.

Your data should be in case-listing form in a spread-sheet with variables in columns and patients in rows. The important variables are related to the outcomes:

(a) disease status (event-local recurrence or distant recurrence or both (all coded as 1) vs. none (coded as zero));

(b) time to event - in months or years (date of event minus date of starting treatment or diagnosis for those with event; for those without events, date of last folloow up minus date of starting treatment or diagnosis).

All other variables are only co-variates.

Any life-table method of survival estimation is preferable: Kaplan-Meier or actuarial.

Please refer to

http://survcan.iarc.fr/survivalchap2.php

for further guidance

There is also a paper in IJROBP 2008 on breast cancer by Shanta et al. Please refer it.

Get the date from the time they are considered cured say after the last cycle of chemo drug and then get the date when the disease symptoms re-occurred (i.e. relapse). Use the sts list or sts graph for KM survivor function and sts graph, cumhaz for NA cumulative hazard function in stata to estimate the overall disease-free survival, I think.

If progression is the only outcome studied, no progression during follow up means no failure, So 100% PFS

to have the value of p for a period of 5 years, should I limit the duration of the follow-up to 5 years and exclude the rest of the patients who have a survival more than 5 years. how can i deduce the 5-year survival estimate through kaplan-meier data and survival tables?

Hi Ihsane, to the best of my knowledge, you can incorporate patients with overall survival(OS) more than 5 years. eg. if the OS is 8 years then code accordingly and you'll get the 8 year survival plot.

Use SPSS, Kaplan Meier option. Ask an epidemiologist for help.

What will be the disease free survival in metastatic cases, who just don't show any response to treatment. In that case should dfs time period be taken as zero and should they also be coded as zero (event not reached) or as one (event reached)?

I am sorry, it's my question and not an answer.