The Short Case for Neoprobe: Skeptical of Its Primary Asset's Success

  • Current Price: $5.25.
  • Price Target: $0.25 by May 31st, 2012.

Neoprobe (NEOP) has 146 million shares outstanding by the treasury method, resulting in a market capitalization of $767 million. With $10 million in cash on the balance sheet, $31 million receivable from equity conversions and $30 million receivable from the sale of their imaging business, Neoprobe has cash of roughly $71 million, resulting in a fully diluted enterprise value of roughly $700 million. The capital structure is slightly complex in this regard, but it is important to know exactly what one is paying for when they trade Neoprobe shares.

Almost all of the value of Neoprobe is in their lymphatic mapping agent, Lymphoseek. This is the only asset I will value. I will endeavor to prove that this asset is close to worthless, and therefore, equity shares of Neoprobe are worthless as well.

Lymphoseek is proposed to be used as a sentinel lymph node biopsy tracer agent. After injecting Lymphoseek, surgeons will be able to detect lymph nodes by scanning the injected area with a radioactive probe. Once a radioactive lymph node is found, it can be removed and dissected to determine if the node contained cancerous elements. The term “sentinel lymph node” refers to the first lymph node that drains from the tumor site.

The current standard of care for sentinel lymph node biopsy is the combination of a blue dye (isosulfan blue or methylene blue) and a radioactive technetium-99 sulfur colloid. This combination results in the identification of a sentinel lymph node upwards of 95% of the time[1]. It is commonly believed that this identification rate cannot be improved upon. There are many studies which prove that the excision of just one or two sentinel lymph nodes accurately predicts the cancer status of the rest of the axilla (armpit).

Obviously, as one removes more nodes, one gets closer to 100% accurate prediction of the axilla status. This concept is called the “received operating characteristic” or ROC curve. It is very important to understand this concept. As one removes and dissects more nodes, one will find more cancer.

The marginal benefit of doing so can be easily plotted on this ROC curve. One can reach 100% sensitivity at the cost of lower specificity. The goal in any diagnostic is to balance the two, if possible. Both blue dyes and radioactive colloids can be interpreted more loosely to remove more nodes—physicians know this, but choose to take just one or two nodes as there is no benefit of doing more as per the ROC curve.

Let’s take a step back from the statistics of diagnostics and learn a little bit about this sentinel lymph node market. It is likely about 200,000 sentinel lymph node biopsies are done for breast cancer in the United States annually[2]. The AWP of sulfur colloid is $53.20[3]. Liberally assuming 300,000 procedures are done, so as to include melanoma and head and neck cancer, the total opportunity for Lymphoseek in the United States is $16 million.

Even if Lymphoseek is priced at a 6 fold increase from sulfur colloid prices (despite no documented advantage, and selling to a cost-conscious hospital market) and takes 100% market share, the fair value of Neoprobe using a liberal DCF is $2.71, a 48% decline from current price levels for an absurd, best-case scenario.

A more likely scenario is Lymphoseek is priced at a 600% premium to sulfur colloid and takes 10-15% United States market share. Any market share gains generally take time. My “best-case” scenario assumed that Lymphoseek takes 100% market share in the first day of approval. Usually when I assume variables like this for other companies, the share price upside is dramatic. This reflects my notion that no matter what happens with respect to market share, Neoprobe is a short.

Pricing is an important lever in this model. If Lymphoseek is priced at $1,000 or more per dose, it still would not generate enough revenue for Neoprobe to be a “buy”. Even the uber-bulls with investment banking conflicts are estimating a price for Lymphoseek of $500. At $5,000 (100 times sulfur colloid pricing), we start to get to a reasonable buying opportunity.

So what should Lymphoseek be priced at? As usual, hospitals calculate risk-benefit when they pay for drugs. The only thing they care about when prescribing medicine is, “is there an extreme medical necessity?” and “will this save the hospital money?” Neoprobe will have you believe Lymphoseek can be injected 15 minutes before surgery, saving the hospital money by reducing wait time and hospital stays. Sulfur colloid can be injected two hours before surgery. Some have been successful using sulfur colloid 30-45 minutes before surgery.[4] This would be a marginal, but useful innovation if the patient had to wait in the operating room.

nfortunately for Neoprobe, injection of colloid and dye take place prior to the patient’s admittance to the operating room[5], either in the surgeon’s office or in the radiopharmacy. In our fantasy world where Lymphoseek easily receives approval, we cannot conjure a realistic scenario where the shares are worth much more than $0.50.

The most likely scenario, should Neoprobe receive Lymphoseek approval, is pricing of 600% that of generically available off-label sulfur colloid, some gradual market share gain (10-30%) due to promotion and newness of the product, which would result in $0.54 per share of value for Neoprobe shareholders, -90% from current levels. However, approval of Lymphoseek is far from certain.

In my eyes, Lymphoseek will not win FDA approval. Neoprobe conducted two pivotal studies of Lymphoseek, NEO03-05 and NEO03-09. Remarkably, the Phase III data for NEO03-05 has differed depending on which Neoprobe presentation one looks at. At an analyst day in December 2010, Dr. Cope from Neoprobe presented data from “n=179 subjects” with a “97% concordance rate with vital blue dyes”. This recapitulated the first report of NEO3-05’s results, which were first reported in May 2009.

However, in January 2011, at the “OneMedPlace Finance Forum”, Neoprobe disclosed a lower concordance rate of 93%[6], with a p-value of 0.04, close the p=0.05 level of failure. The presentation included other deviations from the original data, including simple variables like sample size. For instance, the December 2010 presentation notes n=136 in the ITT population, but the January 2011 presentation notes n=158. How this could happen is well beyond my ability to guess.

Regardless of discrepancies, the trial design of NEO03-05 and NEO03-09 is bizarre. The company selected the primary endpoint of concordance with blue dye, attempting to establish similarity to only half of the standard of care, blue dye plus sulfur colloid. The trial will not inform physicians on how to use Lymphoseek, since it was not compared to the current standard in a non-inferiority or superiority test.

One of the problems that Neoprobe would encounter in running a study against standard of care is that the standard of care cannot be improved. The detection rate for blue dye plus sulfur colloid is over 95%, so the success of the procedure cannot be improved. It would be too time-consuming to do a head-to-head study that attempted to predict cancer recurrence, the downside of a false negative reading.

So instead, Neoprobe chose to compare Lymphoseek to half of the standard of care. The company even had the nerve to claim “superiority” to blue dye by finding nodes that were positive that blue dye did not pick up. Blue dye is not a competitor to Lymphoseek. Blue dye is used in conjunction with a radioactive tracer to assist the surgeon in identifying sentinel lymph nodes.

Furthermore, claims of superiority are meaningless because patients who have one sentinel lymph node taken out that is positive will have all axillary lymph nodes taken out. Neoprobe fails to disclose if these patients were the ones where Lymphoseek “found” their cancer.

Lymphoseek was invented by University of California, San Diego. Despite the glaring conflict of interest, Dr. Anne Wallace from UCSD was a key investigator in the NEO03-05 clinical trials. We believe Dr. Wallace’s site may have enrolled up to a quarter of the patients in NEO03-05.

The FDA will demand a real-world clinical trial for Lymphoseek. While it’s difficult to predict what the design of such a study will be, we suspect the FDA will enforce the “truth standard” requirement is does for all diagnostics. Lymphoseek is a generic sulfur colloid in my mind, with a very poor (and seemingly suspect) Phase III data set.

Because the market for colloids is so small, I don’t see how Neoprobe is worth any more than $50 million, even if FDA approved.

[1] Superiority of Radioisotope over Blue Dye for Sentinel Lymph Node Detection in Breast Cancer. Hayashida et al. Eur Surg Res 2010;44:111-116.

[2] The Sentinel Node Procedure in Breast Cancer: Nuclear Medicine as the Starting Point. Hindie et al. J Nucl Med 2011;52:405-414.

[3] Noridian Medicare Radiopharmaceutical Pricing for 2005. HCPCS Code A9520.

[4] Concordance and Validation Study of Sentinel Lymph Node Biopsy for Breast Cancer Using Subareolar Injection of Blue Dye and Technetium 99m Sulfur Colloid. Kenneth A Kern. J Am Coll Surg 2002;195:467-475.

[5] Kaiser Permanente Circular on Sentinel Lymph Node Biopsy (here - pdf).

[6] Neoprobe corporate presentation.

Disclosure: I am short NEOP. My position and my firm's position in NEOP can change at any time. I will donate at least 50% of my personal NEOP related profits to charity. I agree to be held accountable to this. My average price is near current levels.

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Tagged: , , , Medical Appliances & Equipment
Problem with this article? Please tell us. Disagree with this article? .