TNBC is more aggressive disease with reduced recurrence-free survival.

Enzo Durante

You should look at the STEEP endpoint system for breast cancer research. The choice of endpoints is critical for the interpretation of any analysis.

http://jco.ascopubs.org/content/25/15/2127

Given that there have been no international consensus standards for the definition of survival end points used in clinical trials and supporting literature, Adam Olszewski is to be much thanked for citing, above, the STEEP system, a major step in the right direction, although not on its own enough, and not (yet) adopted by ASCO, NCI or other authorities, and, regrettably, rather inconsistently used - when used at all - in the oncology literature.

In the adjuvant setting the ‘STEEP’ system as introduced by Cliff Hudis at Memorial Sloan-Kettering and colleagues [1] - must reading for any serious researcher - defines the events contributing to survival endpoints, but note that this is solely for the adjuvant setting - that is the reason in fact that no definition is given for progression-free survival (PFS) and other non-adjuvant endpoints. A joint working group of members from both BIG (the Breast International Group) and NABCG (the North American Breast Cancer Group) is currently undertaking a similar but more expansive effort that includes advanced/metastatic setting endpoints and collaborative initiatives (which I am myself involved in) are addressing the issue aggressively at this time [2], for reasons I will make clear below.

Some Survival Endpoint Definitions

But first the definitions. Distilling these proposals (STEEP and BIG/NABCG [private communication]), I present the following proposed working definitions of survival endpoints, relativized for the sake of this discussion to the breast cancer context (to generalize the definition just substitute "disease" for "breast cancer"):

Recurrence-free survival (RFS) includes (1) any recurrence (local or regional [including invasive ipsilateral tumor and invasive locoregional tumor], or distant) and (2) death due to any cause (both BC and non-BC causes of death). Note that under STEEP DFS (disease-free survival) is interchangeable with recurrence-free survival.

Relapse–free survival is defined as any disease recurrence (local, regional, or distant), but death is censored (not included). So recurrence-free survival includes all-cause death which relapse free-survival does not, as seen in the formal definition of relapse-free survival used in the landmark TNBC studies such as Rebecca Dent's: "the time of diagnosis to development of first evidence of clinical or radiographic metastatic disease" [3], among others.

Progression-free survival (PFS) is defined as the time elapsed between treatment initiation and (1) metastatic tumor progression - but note NOT local or regional progression - or (2) death from any cause, with censoring of patients who are lost to follow-up (see [4], and numerous others to same effect). With PFS therefore we are concerned only with distant, not locoregional, disease progression, and with death from any cause (like RFS (recurrence-free survival)).

Finally, time to progression (TTP) differs from PFS solely in that the event of interest is only disease progression, so it does not include death from non-BC causes.

The Problem: Lack of Consensus

However, be aware that even years after the STEEP proposal, there is still wide inconsistency across the literature in the use of these definitions. So for example, 13 of 16 studies recently reviewed [5] counted death as an event in treating TTP, making TTP indistinguishable from PFS, leading the reviewers to conclude that "The lack of uniformity regarding the definition of end points may lead to miscommunication and to confusion when results of different trials are compared, and uniform adoption of the definitions seems therefore in order".

Similarly, in a systematic review [6] of 125 articles that included primary analyses of RCT with survival end points, it was found that clear definitions of the survival end point used were not provided in almost half (48%) of these papers, and 68% reported insufficient information of the survival analysis altogether. And another review [7] demonstrated that whether contralateral breast cancer or non-disease (non-BC) related death was included or excluded in the relevant endpoint definitions significantly affected the estimated outcome probability, making a clinically relevant difference in the conclusions derivable from major trial such as the ATAC Trial (comparing tamoxifen with anastrozole).

Another Plea for Survival Endpoint Standards

This is clearly a wholly unacceptable situation in a medical science that pretends to methodological rigor research integrity, and it is long overdue that standards authorities, as I and others have been calling for years, finally issue comprehensive guidelines for the consistent use of survival endpoints, and that journals aid this initiative in demanding clear and standard definitions for any and all survival endpoints and analyses used in any study submitted for publication.

References

1. Hudis CA, Barlow WE, Costantino JP, et al. Proposal for standardized definitions for efficacy end points in adjuvant breast cancer trials: the STEEP system. J Clin Oncol 2007;25(15):2127-2132.

2. Bartsch R, Dubsky PC, Loibl S, Steger G. Opinions on the ASCO 2011 Annual Meeting. Breast Care (Basel) 2011; 6(4):315-319.

3. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res 2007 Aug 1; 13(15 Pt 1):4429-34.

4. Green S, Benedetti J, Crowley J. Clinical Trials in Oncology. London: Chapman & Hall; 1997. p. 40.

5. Saad ED, Katz A. Progression-free survival and time to progression as primary end points in advanced breast cancer: often used, sometimes loosely defined. Ann Oncol 2009; 20(3):460-4.

6. Mathoulin-Pelissier S, Gourgou-Bourgade S, Bonnetain F, Kramar A. Survival end point reporting in randomized cancer clinical trials: a review of major journals. J Clin Oncol 2008 Aug 1; 26(22):3721-6.

7. Nout RA, Fiets WE, Struikmans H, Rosendaal FR, Putter H, Nortier JW. The in- or exclusion of non-breast cancer related death and contralateral breast cancer significantly affects estimated outcome probability in early breast cancer. Breast Cancer Res Treat 2008; 109(3):567-72.

Constantine Kaniklidis

Director of Medical Research,

No Surrender Breast Cancer Foundation (NSBCF)

European Association for Cancer Research (EACR)

Thank you Constantin for your clearification .... The end point always confusing me and I hope that on the future that the Doktors on the hospitals make it alitel bit clear :)

ENDPOINT MADNESS IN ONCOLOGY

Welcome to the discussion, Tamer. I am already familiar with your work in inflammatory breast cancer (IBC) - among others - a specialization  I share, and I follow the work and publications of NCI/Egypt (I do considerable research on themes of cancer and culture in the  Middle East).

Thanks for your shrewd comment. If only, however, that you and I were in charge of a consensus panel to finally bring long-absent convergence and consistency to the massively inconsistent world of oncology outcome definitions and standardization!

Unfortunately, the situation is more divergent and chaotic than you suggested or we both  would like. So, as  to relapse-free survival (RFS), for instance, the formal definition offered by NCI (US) is "In cancer, the length of time after primary treatment for a cancer ends that the patient survives without any signs or symptoms of that cancer" in which death is censored [1].  This  definition is in turn the basis of the Wikibooks Radiation Oncology/Survival Curve Terminology entry [2], noting "Does not include death (i.e., deaths are censored)", which in turn is adapted from NSABP B-04.

But the matter is worse still: consider in the systematic review of survival endpoints used in phase III clinical trials of adjuvant treatments for colon cancer published from 1997 through 2006 [3], where the relapse-free survival (RFS) endpoint was used in 4 trials, of which RFS was defined as time to recurrence or death only in one trial; as time to recurrence or second primary or treatment-related death or colon cancer specific death in another trial, while in the other two trials using RFS as  endpoint, RFS was defined as time to recurrence (death censored). (And in one study, relapse-free interval (RFI) was defined solely as time to recurrence of any malignancy).

Similarly, three different - and incompatible - definitions of disease-free survival (DFS) across 27 trials were uncovered, and in a most distressing 37% of the trials, absolutely NO definition of any of the outcome endpoints were given! And this in Phase III RCTs, no less! And my own field experience and review since the 2007 publication confirms that the situation has not materially improved, and is true across trials in multiple different malignancies. This confounds making legitimate cross-trial comparisons of outcomes, and allows investigators far too much control over selecting an outcome endpoint, and definition, that best serves their need. 

This is not objective science.

That is why I support encouraging and adopting some reasonably authoritative and standardized expert consensus definitions, in the spirit of the recent DATECAN, the DATECAN initiative's for the Definition for the Assessment of Time-to-event Endpoints in CANcer trials [4] completed for  several malignancy types, and developed by UNICANCER under the auspices of the EORTC (European Organisation for Research and Treatment of Cancer), for all oncological outcome endpoints in clinical trials (despite some serious reservations I retain on some of the proposed definitions).

What I would however still hope, and would work to encourage, is more outreach for input from regulatory and other cancer guideline authorities from non-Western regions like the Middle East, Asia, and Africa.

REFERENCES

1.    NCI Dictionary of Cancer Terms. National Cancer  Institute (NCI). U.S. Department of Health and Human Services. National Institutes of Health (NIH). Available at: http://www.cancer.gov/publications/dictionaries/cancer-terms?cdrid=655254. Accessed: 19 October 2015.

2.    Survival Statistics. Radiation Oncology/Survival Curve Terminology. Wikibooks, a Wikimedia Project. Available at: https://en.wikibooks.org/wiki/Radiation_Oncology/Survival_curve_terminology. Accessed: 19 October 2015.

3.    Punt CJ, Buyse M, Hohenberger P, et al. Endpoints in adjuvant treatment trials: a systematic review of the literature in colon cancer and proposed definitions for future trials. J Natl Cancer Inst 2007 Jul 4; 99(13):998-1003.

4.    Gourgou-Bourgade S, Cameron D, Poortmans P, et al. Guidelines for time-to-event end point definitions in breast cancer trials: results of the DATECAN initiative (Definition for the Assessment of Time-to-event Endpoints in CANcer trials). Ann Oncol 2015; 26(5):873-9.

Does any one know if there is further guidance/consensus for these definitions?

Overall survival = Time to death

Recurence free survival = Time to death or recurecne

Disease free survival = Time to death, recurence or second primary

This is my rudimentary understanding anyway.

Thanks. What about the starting point for Recurrence Free survival and Disease free survival?

Progression free survival is typically used in patients with measurable disease.

When you give treatment, some patients will respond ,some remain stable and some progress.

Jagathnath Krishna K M The start point would be Onset date.

Thanks Shravya Jasti

They all mean the same thing in oncology. The time from treatment (or intervention) to recurrence (which then become censored) or to the last follow-up (or end of observation time) with no recurrence (uncensored).

Ibrahim O. Bello

Salam Alekom

This is my understanding for Standardized definitions for the assessment of time-to-event endpoints in cancer, after screening many papers.

Eman Toraih

Thank you for the standardized definitions. They are instructive but more like fine print. In most cases, recurrence or locoregional metastasis are the only sign of progression especially in tumors that rarely have distant metastasis like my own area (progression, relapse, or locoregional recurrence). I tried to simplify but when one has to be definitive, we have to stick with what standardized definitions you have here