Success will be highly dependent upon selecting the right biomarkers

· TGI - Biomarkers


Elie Dolgin

Nature Medicine 18, 1312–1313, (2012) doi:10.1038/nm0912-1312a

Published online 07 September 2012

The announcement last month that the biotech powerhouse Amgen was terminating a late-stage pancreatic cancer trial added another blow for drug developers hoping to halt tumor growth by blocking the protein known as insulin-like growth factor 1 receptor (IGF1R). In 2010, Pfizer abandoned its IGF1R-targeting antibody called figitumumab after phase 3 trials failed to demonstrate signs of clinical efficacy. And in the two years since, several other drugmakers have discontinued their own earlier-stage programs directed at the same target

With the phase 3 failure of Amgen’s ganitumab, “the field of IGF inhibition is at a crossroads,” says Milind Javle, an oncologist at the MD Anderson Cancer Center in Houston who is running a clinical trial involving the Merck IGF1R blocker dalotuzumab. “This should serve as an example for us that we will not achieve success [with IGF1R inhibitors] unless we identify patients with predictive biomarkers who will benefit. There’s really no alternative. We have to do it.”

A promising potential biomarker that could help parse out patients who would benefit most emerged in July when a team led by Raju Kucherlapati, a cancer geneticist at the Harvard Medical School in Boston, published the largest genome-wide analysis of colon and rectal cancer to date. The group’s analysis of 276 samples revealed that insulin-like growth factor 2 (IGF2)—a hormone that binds IGF1R to promote cell division—was overproduced in tumor samples from 22% of study participants (Nature 487, 330–337, 2012). “We think that all of those 22% of the patients could benefit from the inhibition of either IGF2 or its receptor,” Kucherlapati says.

However, the biomarker approach is far from a sure bet. In a study reported earlier this year at the American Society for Clinical Oncology meeting in Chicago, researchers found that people with colorectal cancers with high amounts of IGF2 responded badly to Merck’s dalotuzumab. By comparison, people with the same tumor type showing elevated levels of a related protein, called IGF1, typically fared better on dalotuzumab than those without the predictive biomarker.

Given the mixed results to date, scientists say that more biomarker discovery trials are needed. Paul Haluska, an oncologist at the Mayo Clinic in Rochester, Minnesota, is leading one such study involving a small-molecule drug from New York’s Bristol-Myers Squibb. In the 100-person trial testing its BMS-754807 in women with breast cancer, Haluska and his colleagues plan to biopsy tissues prior to commencing therapy and then again one month into the protocol to search for molecular signatures the correlate with clinical responses. “Unfortunately, we don’t have many trials like that,” Haluska says. “We’re wasting all this money—and potentially wasting all these novel therapies—by not doing the right thing, which is to identify the patients who are going to benefit.”

Growth-factor growing pains

Some drug companies still working in this space are heeding Haluska’s call. “We believe that success […] will be highly dependent upon selecting the appropriate tumor types and combinations, as well as working with the right biomarkers,” says Tracy Henrikson, a spokesperson for ImClone Systems, a New Jersey–based subsidiary of the Indiana drug giant Eli Lilly that has an IGF1R-targeted antibody called cixutumumab in phase 2 trials.

Yet, others say antibody drugs that target only IGF1R are doomed to fail no matter what, since they hit that receptor but neglect the insulin receptor—another tyrosine kinase receptor that is activated by IGF1 and IGF2 in addition to insulin itself. “There is some evidence, although it has to be proven in humans, that cells can use the insulin receptor when you block IGF1R,” says Stephen Eck, global head of oncology at Astellas Pharma. “So, there’s good reason to believe that inhibiting both would be better than inhibiting one alone.” The Japanese drugmarker has a drug called linsitinib that, like Bristol’s BMS-754807, hits both receptors. The drug is in phase 2 trials for several tumor types—none of which include biomarker-based selection criteria—and “we are continuing our program unabated,” Eck says.

Other strategies under investigation include new combinations of agents or drugs that target the IGF1 and IGF2 proteins rather than the receptors that they bind to. Michael Pollak, a cancer researcher at McGill University in Montreal who has studied the role of IGF1R in cancer for more than 25 years, remains hopeful that one of these various approaches will ultimately pan out in the clinic. “It’s too early to close the door [on IGF1R] because this target has never been evaluated in the modern way,” Pollak says. “Maybe one day the IGF1 receptor will join other failed targets that sounded good but in practice didn’t work, but right now we’re not at that level.”







1 Comment

Comments RSS
  1. larryhbern

    This is interesting. We have been there before, and we don’t learn. IGF1 and IGF2 have structural homology. IGF1 is produced by the liver in response to HGH, but is the most important reason for anabolism. In cancer cachexia, the basic events are similar to stress hypermetabolism and to chronic inanition. IGF1 has a short half-life, similar to TTR. It would be expected to be low. It has been used for kids with short stature, for HIV AIDS, and for burn patients. I won’t comment on cost.

    So I am to understand that IGF2 is overexpressed in GI tumors, and this promotes growth, so you block the IGFR1 to which it binds. But there is a problem because the insulin receptor is not blocked.

    What I don’t see here is a hypothesis that deals with balancing the anabolic and catabolic pathways. If there is less growth from blocking, then what else is seen in the CTrials? I would expect to see catabolic wasting. There may well be effects I don’t expect.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: