[Ip-health] Article on NCI-Sponsored Cancer Trials

[James Love]

I found this report of a study of NCI sponsored clinical trials
interesting.  For example, it claims that perhaps 1/2 of all phase III
oncology trials are publicly funded.  Jamie

--------
* Depending on the eye of the beholder, at least one-quarter and up to
half of new cancer treatments that were tested by the eight
NCI-sponsored cooperative groups' phase III trials turned out to be
beneficial, according to a meta-analysis here.

* Dr. Djulbegovic and colleagues estimated that perhaps half of all
phase III trials in oncology are publicly funded. Of these, the vast
majority are sponsored by the NCI cooperative groups.

* The study was supported by Research Program on Research Integrity,
Office of Research Integrity, and National Institutes of Health. Dr.
Djulbegovic's co-authors reported receiving consulting fees and grant
support from sanofi-aventis, AMGEN, and Eli Lilly.


http://www.medpagetoday.com/PublicHealthPolicy/ClinicalTrials/tb/8878

NCI-Sponsored Cancer Trials Offer Decent Clinical Return on Investment
By Crystal Phend, Staff Writer, MedPage Today
Published: March 24, 2008
Reviewed by Zalman S. Agus, MD; Emeritus Professor
University of Pennsylvania School of Medicine.

TAMPA, Fla., March 24 -- The National Cancer Institute's cooperative
groups seem to have a reasonably good batting average taking treatments
through phase III trials into clinical practice.

-----------------------
Action Points

    * Explain to interested patients that according to this
meta-analysis these publicly funded clinical trials appeared to find
new, better treatments 25% to 50% of the time.

    * Inform patients that clinical trials in cancer are only ethical
when there's about a 50-50 chance that the new treatment will be better
than the comparator.
---------------------------

Depending on the eye of the beholder, at least one-quarter and up to
half of new cancer treatments that were tested by the eight
NCI-sponsored cooperative groups' phase III trials turned out to be
beneficial, according to a meta-analysis here.

The cooperative groups are the Cancer and Leukemia Group B, the Eastern
Cooperative Oncology Group, the Gynecological Oncology Group, the
Northern Central Cancer Treatment Group, the National Surgical Adjuvant
Breast and Bowel Project, the Radiation Therapy Oncology Group, the
Southwest Oncology Group, and the Children's Oncology Group.

New treatments had a better risk-benefit ratio than standard treatment
in 41% of randomized trials from National Cancer Institute cooperative
groups, reported Benjamin Djulbegovic, M.D., Ph.D., of the H. Lee
Moffitt Cancer Center here, and colleagues in the March 24 issue of
Archives of Internal Medicine.

Only 30% of the trials had statistically significant results, but 80% of
these findings favored the investigational cancer treatment over the
standard treatment. New treatments were significantly superior in 24% of
comparisons.

When considering the overall benefit-to-risk ratio, the original trial
investigators determined experimental treatments came out on top in 41%
of comparisons (316 of 766) whereas standard treatments won out in 59%.

The original researchers also declared 15% of the trials to have
revealed major advances in cancer care. Twelve trials -- just 2% overall
-- showed a dramatic 50% or greater reduction in mortality with the
experimental intervention compared with standard treatment.

"Society has received a good return on its investment in the cooperative
oncology group system" that funded the trials, the investigators wrote.
"This pattern of successes has become more consistent over time."

Dr. Djulbegovic and colleagues estimated that perhaps half of all phase
III trials in oncology are publicly funded. Of these, the vast majority
are sponsored by the NCI cooperative groups.

The researchers looked at a consecutive series of all phase III
randomized controlled trials completed from 1955 through 2000 through
the cooperative groups. They found 781 randomized comparisons from 624
trials with a total of 216,451 patients.

Ten percent were unpublished, but the overall methodologic quality was
high.

The majority of trials with nonsignificant findings were inconclusive
(50%) rather than true negatives (3%). Another 19% showed little chance
that the standard treatments were superior to the new comparator whereas
27% were inconclusive but were unlikely to have favored the experimental
arm.

Why so many trials were inconclusive was not entirely clear, Dr.
Djulbegovic and colleagues said.

However, the reasons "appear not to be related to difficulties in the
accrual of patients or other logistical problems," they said, "but
rather to the researchers' overly optimistic assessment of the size of
treatment effects that they designed their trials to measure."

The absolute number of discoveries from these publicly funded trials
might be improved if the number of inconclusive trials could be reduced,
they said.

Over the years, pooled data from the cooperative group trials favored
new treatments for patient outcomes.

Investigational treatments in the trials reduced the risk of death from
any cause by 5% compared with established therapy (HR 0.95, 99% CI 0.93
to 0.98). New cancer treatments reduced the composite risk of relapse,
progression, or death by 10% (HR 0.90, 99% CI 0.87 to 0.93) and
increased response rate 11% (HR 0.89, 99% CI 0.81 to 0.98).

The largest survival advantages were seen in trials treating
gastrointestinal and hematologic malignant neoplasms.

While these benefits generally held up when the analysis was restricted
to only primary endpoint outcomes, new treatments were actually
associated with increased treatment-related mortality (OR 1.43, 99% CI
1.26 to 1.62).

More recent trials tended to be more consistently positive effects with
fewer outliers in either direction, which could have been because the
sample of trials increased over time, the researchers said.

Altogether, it was "ethically a welcome finding" that no more than half
of the trials favored experimental treatments, Dr. Djulbegovic and
colleagues said.

"If every randomized controlled trial found that the new treatments were
better, it would destroy the system of randomized controlled trials"
because patients would not accept randomization, they said.

Although it could be argued that industry-sponsored trials may have a
higher rate of success because of more detailed knowledge of the drugs
under development, industry has been unwilling to share unpublished data
to make the same evaluations, the investigators noted.

Indirect comparison, though, using the proportion granted regulatory
approval suggests a similar 50% success rate for drug development, they
concluded.

The study was supported by Research Program on Research Integrity,
Office of Research Integrity, and National Institutes of Health. Dr.
Djulbegovic's co-authors reported receiving consulting fees and grant
support from sanofi-aventis, AMGEN, and Eli Lilly.

Primary source: Archives of Internal Medicine

Source reference:
Djulbegovic B, et al "Treatment success in cancer: New cancer treatment
successes identified in phase 3 randomized controlled trials conducted
by the national cancer institute-sponsored cooperative oncology groups,
1955 to 2006" Arch Intern Med 2008; 168: 632-642.


--
_____________________________
James Love, Knowledge Ecology International (KEI)
http://www.keionline.org, mailto:james.love@keionline.org
voice +1.202.332.2670, fax +1.202.332.2673, US mobile +1.202.361.3040, Geneva mobile +41.76.413.6584

When everyone thinks the same, no one thinks.  Bill Walton remix of Walter Lippmann