BB&Ts

Sometimes the best information about the performance of medical technologies comes from autopsies. A new study that pathologically assessed drug eluting stent (DES) fractures puts the rate of broken stents at 29%, far above the 1% to 2% previously reported.

First-generation DES stents, longer stents, and specifically the Cypher stent made by Cordis (Miami Lakes, Florida), a unit of Johnson & Johnson (New Brunswick, New Jersey), were all found to be more apt to break than the Taxus DES stent made by Boston Scientific (Natick Massachusetts). “If I were a patient, I would ask 'Why are you using Cypher when there are better stents on the market?' I wouldn't let them put either the Cypher or Taxus into me,“ said the study's principal investigator Renu Virmani, MD, president and medical director, CVPath Institute (Gaithersburg, Maryland).

Virmani and colleagues studied high-contrast film-based radiographs of 177 consecutive lesions from a CVPath DES autopsy registry. Stent fractures were graded on a scale of I to V and adverse pathologic findings (thrombosis and restenosis) were assessed histologically. “Fractures usually occur when the artery was kinked or bent and then straightened when the DES was implanted,“ Virmani told BB&T. “Those sites have excessive stress and the stress is usually in the middle. And if it's overlapped, the incidence is even higher. When you overlap, there is more strength, but where the overlap finishes, it will fracture.“

Specifically, in overlapping stents, most fractures were observed within a 5 mm distance from the overlapping zone. Stent fracture was documented in 51 lesions. Longer stent length, use of Cypher, and longer duration of implant were all identified as independent risk factors of stent fracture. “Most of the fractures in the Cypher stents were located in flexible N-shaped, undulating longitudinal intersinusoidal-ring linker segments, whereas in Taxus Express stents, fractures were observed in the straight longitudinal inter-crown linker or the modular ring portion,“ Virmani and colleagues wrote in the current issue of Journal of the American College of Cardiology.

Stent fractures can cause thrombosis (blood clots) and restenosis (a narrowing of the blood vessel, restricting the flow). Virmani found that although slight fractures didn't seem to have adverse effects on patients, almost three quarters of the fractured stents studied revealed signs of scar tissue or blood clots. “Of those 51, we had nine cases with severe fracture, meaning the stent is broken through and through,“ she said.

As an example, the team cited a case of a 60-year-old man with a grade V Cypher stent fracture who reported having chest pain 60 days after stent placement and underwent angiography. Stent thrombosis was found and treated by balloon angioplasty followed by bypass surgery. But the patient later died of undisclosed surgical complications.

Another man, 58 years old, died of subacute stent thrombosis 11 days after implantation with the Cypher stent. In that case there was a complete separation of the Cypher stent distal to the site of overlap with thrombotic occlusion. At the site fracture, the stent was under expanded, and a blood clot was observed in the vessel. Virmani pointed out that her team studied first-generation stents, but that some of the newer generation DESs are not yet available in the U.S.

“The strength of stents has to change and the design has to change,“ she said. “The Cypher is an old stent. Now they've got Liberté. The rest of the world has newer stents but not the U.S. Ask the FDA why we are so slow.“

Boston Scientific's Taxus Liberté Atom paclitaxel-eluting coronary stent system was FDA approved in May. It's the second generation of the Taxus Liberté and is a little smaller than most of Boston Scientific's other DES offerings. Virmani believes this is the first report of stent fracture in a large series assessed at autopsy with a highly sensitive method. “We've done what we had to do; now we have to look at these next-generation stents. I want the public to be better informed,“ she said.

Despite economic environment, new ideas needed

Fittingly enough, given the locale of November's MidAmerica Healthcare Venture Forum, the East and West Coast-based members of a panel on “Coast-to-Coast Healthcare Venture Capital Investing“ showed an interest in steering some of their investment dollars to Midwestern start-ups.

Last month's session at the Monona Terrace Conference Center alongside the sun-splashed waters of Lake Monona in downtown Madison brought the requisite acknowledgements that VC investing in healthcare and virtually every other sector is at a low ebb right now, but signs of optimism were commonplace. Rod Altman, senior partner at CMEA Ventures (San Francisco), said that the cost structure for companies operating in the Midwest – as in the Southeast and South Central region – is less, “so opportunities exist there“ for increased VC investment.

John Fletcher, founding partner of Fletcher Spaght Ventures (Boston), said that data gathered by his company identified Michigan and Ohio in particular as regions showing promise for healthcare investing, “so we want to look more enthusiastically there – it has good potential.“ He added, “We like meeting people who have good ideas.“

David Lowe, a partner in Skyline Ventures (Palo Alto, California), said his firm is “agnostic to geography,“ adding, “it is a question of people and ideas.“ Moderator John Neis, managing partner at Venture Investors (Madison), noted that the latest available data on VC investing have not been kind. Figures for the third quarter ended Sept. 30 showed a sharp falloff to a total of $4.8 billion in venture investment, down from $7.2 billion in the prior-year quarter, which came just before the plunge off the economic cliff that is statistically marked as of 4Q08.

Neis, who heads that firm's healthcare practice, observed that the psychology being followed by VCs generally “appears to be that they are hoarding their cash for deals they already are in. There are companies out there getting financed,“ he said, “but they need to be hitting the ball out of the park“ insofar as product development and clinical results are concerned. His observation about portfolio companies ruling the roost when it comes to receiving financing generated a range of responses.

Lowe said, “The deal flow actually is very good for us – it probably has never been better.“ But he added that getting deals together “seems harder, most likely because uncertainty is higher.“ He said the percentage of healthcare deals his firm is getting involved in is skewing more toward device companies than in the past, now amounting to more than 50% of all deals, “because with devices, you can see a way through to taking it to market.“

Altman said his firm is doing “a couple fewer deals,“ and acknowledged that in the current economic environment, “sure, we're putting more money into existing deals.“ He said that because of both economic and political uncertainties, such as the don't-yet-know status of healthcare reform, “there's more work to do at the board of directors level with these portfolio companies.“

Ashley Friedman, senior associate with Investor Growth Capital (New York), said, his firm has been “a bit surprised“ at the difficulty in getting deals done, sometimes even in quoting on possible deals. “You'd think it would be easier to do deals in this environment,“ he said, “but it isn't.

“We have seen deals tip over at the altar because the terms or something else wasn't right.“ Fletcher explained the general trend to shy away from new deals in favor of existing pieces of a portfolio: “You need to have enough money to take a company all the way through to a deal,“ likely an exit via acquisition. “We do in fact tend to put money into a company longer“ than might have previously been the case.

Altman noted that VC investors are “looking more closely at what stage [a company] might be at“ in relation to what the final cycle might be.

Referencing a specific deal, but one that serves as an illustration of the market in general, Lowe said that from a practical sense, there really was only one way to turn. “Absent an IPO market and with the uncertainty in the economy, combining venture capital with a corporate strategic investment was the right deal.“

Friedman said his firm's competitors today “now are not so much other VC firms, but rather, corporate investing.“ As to whether there are more limited paths open to VCs in these troubled economic times, Altman said, “In this market, it's all about survival – the main goal is to get a company financed.“

Citing what he termed “the differential in VC return rate,“ Fletcher said, “in the absence of a public-market option, you have to put a cap on your investment and clearly see a company's public value.“ All of the panelists cited the importance of being involved with other investors in syndicating a financing deal.

It's a matter of making a given company valuable to potential public investors and eventually to possible corporate acquirers. “We have to provide the incentive to get others investors in,“ Altman said. Fletcher added, “In syndicates, each participant is putting in less, but it's still a way to spread out the risk.“

The 7th annual MidAmerican Healthcare Venture Forum was sponsored by the Mid-America Healthcare Investors Network (St. Louis) and conference organizing firm International Business Forum (Massapequa, New York).

Obesity costs could quadruple to $344B by 2018

If current trends continue, 43% of U.S. adults will be obese and obesity spending will quadruple to $344 billion by 2018, according to a new study. On the other hand, if obesity rates hold steady, the U.S. would save nearly $200 billion in healthcare costs – or about $820 per adult.

Based on research by Emory University (Atlanta) healthcare economist Ken Thorpe, PhD, executive director of the Partnership to Fight Chronic Disease (PFCD; Washington), the report is the first to estimate obesity prevalence and costs at the state and national level 10 years from now. The report was commissioned by United Health Foundation (Minnetonka, Minnesota), Partnership for Prevention (Washington), and the American Public Health Association (Washington) in conjunction with their annual America's Health Rankings report.

“At a time when Congress is looking for savings in healthcare, this data confirms what we already knew: obesity is where the money is,“ Thorpe said. “Because obesity is related to the onset of so many other illnesses, stopping the growth of obesity in the U.S. is vital not only to our health — but also to the solvency of our healthcare system.“

The PFCD said it believes that a top priority must be addressing the obesity epidemic through meaningful, evidence-based approaches, including: removing barriers and empowering Americans to take control of their health; educating Americans to see being obese as a serious medical condition that significantly heightens their risk for other health problems; ensuring that fear about the stigma of obesity does not eclipse the need to combat it; redesigning our healthcare system to treat obesity like a preventable medical condition; and engaging employers and communities to get them invested in promoting wellness.

“I am hopeful that health and wellness programs to prevent obesity and other chronic conditions will remain a top priority in final legislation as health reform moves through Congress,“ Thorpe said. “These are problems we can no longer afford to ignore.“

The report projects that obesity will surpass 50% of the adult population in six states, with an associated increase in health spending linked to obesity of more than $1,600 per person in each of these “worst“ states: Kentucky, Maryland, Mississippi, Ohio, Oklahoma, and South Dakota. Oklahoma is expected to have the highest obesity rate in the country by 2018. The report projects that the state's obesity rate will be 56.1% by then, and that obesity-attributable healthcare spending will be $1,906 per adult. If obesity levels remain steady, however, it would provide a potential savings of $1,200 per adult or a savings of more than $3.2 billion for the state.

According to the report, the obesity rate will stay below 35% in only four states and the District of Columbia. Nevertheless, the data shows that obesity-attributable health spending will climb to more than $800 per person by 2018 in each of these “best“ states: Colorado, Connecticut, Massachusetts, Virginia, and the District of Columbia. Colorado is estimated to have the lowest obesity rate by 2018 – 29.8%. Obesity-attributable health spending in Colorado is expected to be $864 per adult, according to the report.

According to the report, Thorpe and colleagues from Emory developed an economic model to estimate the growth of healthcare costs over time that are attributable to changes in obesity rates. They used nationally representative data on adults to estimate the effect of the increasing prevalence of obesity on total direct healthcare costs. Estimates are controlled for age, gender, race, ethnicity, marital status, education, income, health insurance status, geographic region and smoking status, the report notes.

Obesity is growing faster than any previous public health issue our nation has faced, Thorpe and colleagues noted in the report. If current trends continue, 103 million American adults will be considered obese by 2018, the report said. The report also says that obesity-related direct expenditures are expected to account for more than 21% of the nation's direct healthcare spending in 2018.

The rise in the prevalence of adult obesity has been well documented over the last 20 years, according to the report, increasing from 12% in 1989 to 27% in 2008. Thorpe and colleagues noted that the true prevalence of obesity is likely to be substantially higher, however, as it has been shown to be under-reported by about 9.5%, because people tend to understate their weight on telephone surveys.

FDA seen as obstructionist on FFDM clearance

The saying goes something like this: “If at first you don't succeed, try, try again.“ However, some who attended the Nov. 19 hearing of the radiological devices advisory committee might have been inclined to tell FDA, “If you try again and you still don't succeed, you're fired.“

Whether that was consciously the message, it certainly was nearly the tone of some of the comments aimed at the agency in the hearing, which was held to address the 2008 guidance for 510(k) applications for full-field digital mammography (FFDM) systems. The agency was on the receiving end of a number of unflattering remarks during the open public hearing, including a comment that FDA's decade-plus failure to come up with a 510(k) path for FFDMs “is shameful.“

However, the final vote of the panel indicated that the panelists were unfazed by the views of the experts who addressed the panel despite the fact that among the speakers was the leader of the largest-ever study of digital mammography, the DMIST (Digital Mammographic Imaging Screening Trial) study.

The outcome of the meeting dovetails with the recent announcement by the U.S. Preventive Services Task Force that women between the ages of 40 and 50 should not be automatically screened for breast cancer (see accompanying article), but both developments seem headed for a showdown with American women, who are ill-disposed to take anything for granted where breast cancer is concerned.

FDA's latest guidance on 510(k) applications for FFDM systems, a technology that has been in play for the better part of two decades, came out in May 2008 and followed by two years a previous meeting of the panel during which panelists recommended that the agency regulate FFDM systems as class II devices with special controls. Some firms took a run at the 510(k) route in the 1990s, but those efforts fell apart because studies indicated too much variance between readers of the films. All the same, the three-plus years since the 2006 panel meeting during which the panel made the recommendation helped fuel accusations that FDA is dragging its feet.

One of the agency's critics was Etta Pisano, MD, of the University of North Carolina School of Medicine (Chapel Hill, North Carolina), who led the parade of observers that offered a generous portion of blowback aimed squarely at FDA. Pisano's standing on the issue was especially solid inasmuch as she served as the lead investigator of the DMIST trial.

Pisano remarked that the DMIST trial was more than sufficiently exhaustive, enrolling nearly 50,000 patients to test digital mammography against its analog antecedent, film mammography. Pisano said, “these results found significantly improved sensitivity“ in the study arm, and while she acknowledged early skepticism regarding digital technology, “the time for skepticism has passed,“ she said.

Pisano asserted that FFDM “was developed specifically developed to address the problems with film“ and was in part the result of a push on the part of the National Institutes of Health commencing in 1992 to improve mammography.

“FDA was there for the design of DMIST,“ Pisano said, “along with many other stakeholders.“ She said she earlier supported the necessity of a PMA because of concerns about spatial resolution, but remarked that “the draft guidelines of last [year] would be unethical today“ because of the implicit need for double exposure to X-radiation. “It is my considered judgment as a private citizen that any study that requires double exposure ... should not be IRB approved.“

“Clinical studies are not only not needed but, in my mind, are unethical,“ Pisano said, adding that because digital systems are widely available, “it is going to be extremely difficult to enroll“ enough patients in a trial that calls for randomization to film.

As is well known, the DMIST study completed enrollment of almost 50,000 women in November 2003 to test FFDM against film mammography, and resulted in essentially identical overall rates of detection of suspicious neoplasms. One of the long-standing arguments in favor of FFDM is that the technology cuts down on exposure to x-radiation by as much as 15%, but some estimates are that the reduction runs as high as 25%. However, the technology scored better in women between the ages of 40 and 50 and also did better in detecting abnormalities located in relatively dense breast tissue.

Margarita Zuley, PhD, spoke on behalf of the American College of Radiology (ACR; Reston, Virginia) and the Society of Breast Imaging (Reston, Virginia), argued to the panel that the approach embodied in the 2008 guidance is more burdensome than the least burdensome approach.

Zuley maintained that ACR's phantom test for mammography is more than adequate to determine whether a new product is equivalent to a predicate in clinical terms. She also asserted that FDA has available to it a raft of data from across the globe.

DMIST “was a wonderful study,“ Zuley said, but “there are many other studies out there“ that have collectively captured data from more than 3.4 million women from across the globe. “FFDM has performed at least as well as film“ in those studies, and film mammography offers “no difference in positive predictive value,“ she said.

“We already have all the evidence available to us“ to back equivalence, Zuley continued, adding that “the safety and effectiveness“ of FFDM technology has been established.

As for the differences between screening and diagnosis, Zuley observed that “if we look at all the studies I've presented ... they all had at least one year of follow-up,“ which demonstrates technological equivalence beyond any reasonable doubt.

“If we persist in guidances that limits new products being introduced into the U.S.,“ Zuley said, the consequences will be higher costs, triage, barriers to access in rural areas, “and limiting the ability of vendors to produce better processing algorithms.“

FDAer asserts clinical testing essential: Robert Smith, MD, of the radiological devices branch at CDRH, said he did not believe that the question of the day was whether a 510(k) can be cleared for an FFDM system by means of physical lab testing, with or without phantom testing. “I believe that the question is what quantity and quality of clinical performance does FDA need“ to establish equivalence in both screening and diagnosis.

“Breast cancer is the most commonly diagnosed cancer and the second leading cause of cancer-related deaths among women in the U.S.,“ Smith acknowledged, but nonetheless held that “the relationship between image quality and lesion detection needs thorough analysis and further studies.“

Despite the fact that DMIST employed four FFDM systems, Smith alleged that DMIST taught nothing “about how to compare two different FFDM devices.“ He said that more than 8,700 accredited facilities are operating 43more than 12,000 machines, and that “approximately 59% of the accredited machines are FFDM“ units.

Smith also cited the 2001 PMA guidance for FFDM, which he noted includes the statement that “a clear correlation between design/bench parameters and clinical performance has yet to be established.“

In the end, the panel concluded that clinical data are needed to demonstrate substantial equivalence unless both the software and the hardware in the predicate and tested device are identical or nearly so. The conclusion was based on the use of evaluation standards set by ACR in 2007, but a formal consensus was not reached by the panel on the number of cases needed to establish equivalence. In any event, two cases were seen as insufficient, but a couple of panelists were of the view that 30 cases are more than would be needed in most studies.

However, the discussion included a series of remarks that suggested that if clinical data are needed, a sponsor might have to enroll as many as 4,000-5,000 subjects to sufficiently power a study of whether the study article offers a similarly reliable view of the general features of a patient's breast tissue. A multiple-reader/multiple-case study might shave these numbers down.

On the question of whether stress testing is necessary in a clinical evaluation of a system, the panel concluded that in many instances it would not be, but that when stress testing is desirable, it should include a check on detection of masses with any one of several characteristics, including microcalcifications of diameters of 300-500 microns.

Recommendation odd backdrop to meeting

The timing of the meeting of the radiology devices advisory committee fell precisely one day after the U.S. Preventive Services Task Force (USPSTF) announced its recommendation regarding breast cancer screening. The announcement by USPSTF stated that breast cancer screening for women between the ages of 40 and 49 should not be universal, but rather “should be an individual one and take patient context into account, including the patient's values regarding specific benefits and harms.“

The announcement was not well received, to put it mildly, and reverses a policy announcement made by USPSTF earlier this decade that led to annual screening for this population. The oscillation on this issue seems to mirror the question of how aggressively – and even whether – to treat men for high levels of prostate-specific antigen and/or prostatic hyperplasia, another medical question that has generated enormous controversy.

According to the Breast Imaging Commission, part of the American College of Radiology (Reston, Virginia), the adoption of that recommendation could reverse “two decades of decline in breast cancer mortality“ and result in “countless American women [dying] needlessly from breast cancer each year.“ The BIC statement also notes that the guideline was crafted by “a federal government-funded committee with no medical imaging representation“ and that the announcement comes across as “a conscious decision to ration care.“

The American Cancer Society (Atlanta) was no more enthusiastic than others about the move. ACS's Nov. 16 statement notes that “the most recent data show us that approximately 17% of breast cancer deaths occurred in women who were diagnosed in their 40s, and 22% occurred in women diagnosed in their 50s.“ ACR acknowledged “the limitations of mammography,“ adding that the association “remain[s] committed to finding better tests,“ and is currently “funding a large study to improve the accuracy of mammography.“ The statement, however, makes no mention of the nature of the study.

CMS to eye pharmacogenomic testing for cancer

The Centers for Medicare & Medicaid Services will convene a meeting of the Medicare Coverage and Evidence Development Advisory Committee (MedCAC) in January 2010 to examine the evidence behind pharmacogenomic testing for cancer in an attempt to determine the need for a coverage determination.

The Nov. 25 announcement states that the meeting, scheduled for Jan. 27, will review the evidence for genetic testing for five gene/treatment pairings, two of which are for breast cancer treatments. One of these is for the use of tamoxifen in relation to the CYP2D6 gene (one of the cytochrome P450 genes), and the second is the use of trastuzumab for the HER2/neu (human epidermal growth factor receptor 2) gene.

The three other tests involve irinotecan for colon cancer (UGT1A1), imatinib for chronic myelogenous leukemia (BCR-ABL) and cetuximab for colorectal cancer (K-ras).

CMS acknowledges in the announcement that “evidence from clinical studies suggests that tumor genetic factors may be significant markers or predictors for a tumor's development and sustained growth“ as well as for a tumor's “likely response to certain anti-cancer agents.“ The statement also notes that the agency “is aware that the body of evidence on the role of pharmacogenomic testing in cancer continues to evolve.“

Michael Stocum, managing director of Personalized Medicine Partners (Raleigh, North Carolina) told BB&T that the majority of these scenarios are at least fairly well understood. “For most of them, it's strongly fleshed out,“ he said. Regarding the use of imatinib for chronic myelogenous leukemia (CML), he said, “The very mechanism that drives CML can be translocated“ with the drug, which he said “markedly drives outcomes.“ He posed the question, however, of whether other factors have to be addressed, including whether other genes ought to be checked and whether messenger RNA should be examined.

As for the use of tamoxifen in breast cancer, Stocum, who chairs the personalized medicine advisory group at the American Association for Clinical Chemistry (Arlington, Virginia) said, “this has been pretty well published.“ Of the group, he said the case for irinotecan for colon cancer is the spottiest, but that the others are well supported by existing evidence.