) -- Wall Street research is separate from investment banking. Sell-side analysts spend their days analyzing companies and counseling investors on which stocks to buy and sell, while their investment banking colleagues, oblivious to what the analysts might be writing or saying, raise money and negotiate deals for those same companies.

Just kidding.

Experienced investors understand that, despite regulatory overhauls, Wall Street research is still fundamentally flawed and rife with conflicts. Investment bankers always win, which is why investors who rely heavily on sell-side research and don't do independent work too often lose.

In January, Boston-based


(VSTM) - Get Verastem, Inc. Report

raised $59 million in an initial public offering led by UBS and Leerink Swann, with Rodman & Renshaw, Lazard, and Oppenheimer lending secondary support. Biotech IPOs are relatively rare these days, but Verastem's offering was even more unusual and risky because the company's drug development program is based on research into cancer stem cells still in preclinical stages. Verastem won't initiate Phase 1 studies, which test a drug's safety in humans, until next year. Proof-of-concept efficacy data will not available until late 2014 at the earliest.

Biotech companies with drugs in much later stages of clinical development find it difficult to go public today, yet here was Verastem, with nary a single patient exposed to any of its drugs, selling 5.5 million shares to the public at $10 per share.

Forty days later, the minimum time period allowed by law, sell-side analysts from all five of the investment banks which took Verastem public issued glowing reports with buy ratings and price targets 50% to 100% above the current share price.

Coincidence? Not likely.

The truth is that major investment banks are still hopelessly biased in favor of the companies with which they do business. Analysts working for those investment banks, therefore, have strong inducements to issue bullish research.

The Verastem IPO, while modest, still generated $4.4 million in banking fees. To partake in the lucrative business of raising capital, a bank can't alienate management with even the risk of a negative review.

I'm going to dissect the Verastem research report issued by UBS for two reasons. First, UBS is by far the highest-profile bank in the Verastem syndicate (a term for the group of banks that orchestrates an IPO.) Second, the analyst who wrote the report, Matthew Roden, is a smart guy who often does quality work and has made good calls in the past. My complaint isn't with Roden per se, but rather with the "reformed" research-investment banking system that still creates misaligned incentives.

Before I demolish Roden's Verastem research (nothing personal, Matt), a bit of background on Verastem plus an explanation of how a professional investor values development-stage drug candidates.

Verastem was founded on the belief that most currently marketed anti-cancer drugs have little efficacy against cancer stem cells (CSCs), a progenitor subgroup of cells that some scientists believe fuel metastases and tumor regrowth.

The company's first drug candidate is VS-507, known chemically as salinomycin. A member of the polyether ionophore class of antibiotics, scientists first synthesized salinomycin in the late 1960s; it's currently used to treat feedlot animals. Based on data collected in the laboratory, Verastem plans to focus on "triple negative" breast cancer (TNBC), a patient subgroup whose tumors have adverse biological characteristics that renders current therapies particularly ineffective.

As I said above, Verastem won't start dosing humans with VS-507 until next year. The company's biological hypothesis itself -- I will spare you the details -- has no clinical validation whatsoever and the company's development program lags at least eight CSC-focused competitors, including


(NVS) - Get Novartis AG Sponsored ADR Report






(PFE) - Get Pfizer Inc. Report

and privately held

OncoMed Pharmaceuticals

. Many of these competitors already have clinical data in hand. As a decades-old drug, salinomycin also has no composition-of-matter patent; Verastem intends to file for less compelling formulation and method-of-use patents, but has not yet done so.

To account for the risk of drug development failure, investors routinely use a subjective adjustment variable known as probability of success (POS).

If you're bothered by the childishly funny acronym, you can use "probability of technical success," or POTS. Either way, a POS adjustment is a risk adjustment applied to each molecule in a company's pipeline that takes into account the compound's class, target indication, available data, and -- crucially -- stage of development.

TheStreet Recommends

Preclinical drug candidates are generally assigned a 10% POS, which means that for every ten preclinical-stage drugs beginning development, one will make it all the way to approval. A 10% POS may seem low but it actually overstates R&D success rates. From 2004 through 2011, the biotech industry reported a 6.7% success rate for cancer drugs entering clinical development for any indication, according to data recently published by the Biotechnology Industry Organization (BIO).

Here's why understanding POS is critical: UBS, in its Verastem initiation report, applied a 30% POS adjustment for Verastem. In other words, Roden, the UBS analyst, believes Verastem's preclinical drugs are 3-4 times more likely to reach approval than the industry norm. This dramatically higher POS adjustment drives all of the upside for UBS's $20-per-share price target on Verastem.

If UBS had used a standard POS for Verastem, the stock would be fairly valued at current levels. But if Verastem were fairly valued at around $11 per share, UBS, Verastem's lead banker, would not have been able to initiate research coverage with a crucial buy rating.

Here's how UBS justifies a 30% POS for Verastem's lead drug VS-507 (salinomycin):

1. "TNBC is a targeted indication." Great, but who cares? Whether it's going after a targeted indication or not, Verastem is still dependent on a biological hypothesis -- cancer stem cells -- that has not been validated. That's the key point.

2. "

Verastem uses Companion diagnostic and biomarker analysis." Companion diagnostics and biomarker analysis are becoming increasingly important for drug R&D, agreed, but patient selection and analysis metrics matter only if the basic hypothesis -- that salinomycin kills CSCs -- is valid. Further, even if these bells and whistles justify a higher POS, it's hard to say how much without data. I'm guessing it's a couple of percentage points, at most.

3. "Recent approvals show validity in approach." This again assumes a valid hypothesis, and might be rephrased as follows: "If the drug works, it will receive approval." Although the FDA has been supportive of applications underpinned by novel biological rationales, historical data clearly show approval risk to be a minor overall contributor to POS at this stage. Reduced approval risk doesn't justify significant deviation from historical norms.

4. "Development plan offers several shots on goal." Allow me to translate: "If salinomycin fails, Verastem has other molecules in the pipeline." This fact is unrelated to the value of the lead program. Further, those other shots on goal, VS-4718 and VS-5095, were licensed out of the utility closet at Poniard Pharmaceuticals -- hardly a Nobel Prize-worthy pedigree. Remember another rule-of-thumb: When a company's lead drug candidate blows up, investors' reaction is always the same: sell first, ask questions and rationally assign value to earlier-stage molecules later. Much later.

The Verastem bull thesis is actually quite simple and doesn't require UBS' 43 pages of gobbledygook:

Verastem's CEO Christoph Westphal is the proverbial ice cube salesman in the Arctic. He's shown an uncanny ability to sell speculative and risky biotech ventures to clueless and gullible pharmaceutical companies for premium prices. I'm referring to Westphal's

Sirtris Pharmaceuticals

and its age-defying resveratrol drugs which


(GSK) - Get GlaxoSmithKline plc Sponsored ADR Report

bought for $700 million. Sirtris has been an embarrassing flop for Glaxo.

Westphal was also a backer of

Alnara Pharmaceuticals

, which recycled a failed pancreatic enzyme replacement from the trash heap that was

Altus Pharmaceuticals

and then sold it to

Eli Lilly

(LLY) - Get Eli Lilly and Company (LLY) Report

for $180 million. Lilly submitted the Alnara drug for FDA approval but was soundly rejected.

Investors bullish on Verastem are hoping Westphal can work his magic again. At least my explanation of the Verastem bull case is honest.

One final point: an hour of research led me to an interesting tidbit about salinomycin not mentioned anywhere in the UBS opus.

A tip from a biotech investing pro: Look for and use a molecule's chemical name, not the company-assigned codename when during your research. It often leads to obscure-but-crucial information. It turns out that salinomycin can only be given safely to some animals -- even small doses kill certain species -- and the compound has a narrow therapeutic window (a term for the range of doses at which a drug is both effective and safe.) Salinomycin's toxicity is due partly to a negative impact on cellular calcium concentrations.

Uh oh.

You don't have to be a Nobel Laureate to know that proper functioning of the heart relies heavily on the precise movement of calcium ions. Inadvertently disrupt this dynamic and there are often dire consequences.

I'd say salinomycin's POS rate today is much closer to10% -- if not lower. With this level of outsized risk and uncertainty, Verastem is more than fairly valued. Is a buy rating justified? Not a chance, unless you happen to be a sell-side analyst who needs to keep his investment banking overlords happy.

Disclosure: Sadeghi-Nejad has no position in Verastem or any other stocks mentioned in this column.

--Written by Nathan Sadeghi-Nejad in New York.

Nathan Sadeghi-Nejad has 15 years experience as a professional health-care investor, most recently as a sector head for Highside Capital. He has worked on the sell side (with independent research boutiques Sturza�s Medical Research and Avalon Research) and the buyside (at Kilkenny Capital prior to Highside). Sadeghi-Nejad is a graduate of Columbia University and lives in New York. You can follow him on Twitter @natesadeghi.