After Hours Summary: NCNO -25.7% down sharply on earnings/guidance; TTEC +20.2% higher as it's ready to engage on takeover offer
After Hours Gainers:
Companies trading higher in after hours in reaction to earnings/guidance: JNJ +0.1% (updates financial impact from imminent closing of its Intra-Cellular Therapies acquisition)
Companies trading higher in after hours in reaction to news: TTEC +20.2% (board completes review of takeover proposal by founder/CEO; co remains ready to engage with Mr. Tuchman), DOMO +3.7% (partnership with Weather Trends for real-time sales forecasting), ASAN +1.4% (CEO bought 449507 shares), CBAN +0.7% (acquires The Ellerbee Agency), GLOB +0.4% (files mixed securities shelf offering), LINE +0.2% (acquires three warehouses), WTM +0.2% (acquires a majority interest in Enterprise Electric), UMH +0.1% (increases dividend)
After Hours Losers:
Companies trading lower in after hours in reaction to earnings/guidance: NCNO -25.7% (also authorizes new $100 mln share repurchase program)
Companies trading lower in after hours in reaction to news: AJG -2.8% (acquires Tresidder Insurance Brokers), AVAV -1.2% (shareholders approve acquisition of BlueHalo), VZLA -0.7% (files for $600 mln mixed securities shelf offering), ED -0.3% (files mixed securities shelf offering)
OpenAI raises $40B at $300B post-money valuation
OpenAI on Monday announced that it closed one of the largest private funding rounds in history.
According to a blog post on the company’s website, OpenAI raised $40 billion in a round that values the company at $300 billion post-money. SoftBank led the financing, CNBC reported. Other participants included Microsoft, Coatue, Altimeter, and Thrive, all of which are earlier backers in the outfit.
“[This new capital] enables us to push the frontiers of AI research even further, scale our compute infrastructure, and deliver increasingly powerful tools for the 500 million people who use ChatGPT every week,” OpenAI wrote in the blog post. “We’re excited to be working in partnership with SoftBank Group — few companies understand how to scale transformative technology like they do.”
CNBC, citing a source familiar with the matter, says that around $18 billion of the funding will go toward OpenAI’s ambitious Stargate infrastructure project, which aims to establish a network of AI data centers around the U.S.
Silicon Valley of energy trading’: the Danish companies seeking an edge in renewables
New breed of traders in Aarhus and Aalborg are using algorithms to navigate volatile and rapidly growing market
The trading floor of InCommodities, on a commercial estate in the Danish city of Aarhus, shows off a dynamic, new side of Europe’s energy markets.
Youthful traders at the company, part-owned by Goldman Sachs, keep their eyes on banks of huge screens, watching for the perfect moment to buy or sell electricity. Other staff are developing algorithms that can help them to do the job far more efficiently than a mere human brain.
All are pursuing an edge in the increasingly complex renewable energy market that has developed as wind and solar generation have grown at the expense of fossil fuels. A mix of high potential profits, intellectual challenge and a role in the transition to a green economy have drawn dozens of smart young trading entrepreneurs to set up ventures both in Aarhus and Aalborg, just over 100km away. They compete with established players such as Equinor’s Danske Commodities in an area that Daniel Andersen, InCommodities’ chief executive for Europe, called the “Silicon Valley of energy trading”.
Andersen said after showing off the trading floor that the expansion of renewable energy generation would increase demand for the services of companies like his.
He added that the trading had a “stabilising effect” on the naturally volatile market, removing bottlenecks in the distribution of electricity.
“It’s incredibly challenging and incredibly fun,” Andersen said. “People get motivated by it.”
That volatility reflects the market’s reliance on changing natural conditions and fluctuating demand. It can move for reasons as small as the strength of the sun on a solar panel in Madrid, the angle of a wind turbine in the North Sea or extra orders at a factory in southern Germany.
The greater complexity and vast amounts of data have given sophisticated trading algorithms far greater importance, creating opportunities for traders such as InCommodities that can design and deploy them.
Andersen said increasing renewable generation would generate growing amounts of data.
“You need to be able to cope with this increase in data,” he said.
However, regulators are also starting to pay closer attention to the rise of automated trading. In a report last year, the Netherlands Authority for Consumers and Markets (ACM) said that while the trend brought benefits such as greater liquidity, it also came with risks. These included greater volatility, lower transparency and unintended market manipulation, it said.
Ian McGowan, head of compliance at InCommodities, accepted there was scope for greater understanding of how algorithms interact with one another.
“With greater quantities [of algorithms], the risk of inadvertent market conduct increases,” he said.
The new market’s emergence reflects how Europe’s energy generation has been transformed. Less than 10 per cent of the continent’s energy came from renewable sources 20 years ago, while the current figure is higher than 25 per cent. Spot prices for energy on the market in Germany — the continent’s biggest power consumer — can swing from negative on days when there is too much power to more than €600 per megawatt hour on days of “Dunkelflaute” when the sun is hidden and the wind is light.
Mads Schmidt Christensen, vice-president of strategy at Danske Commodities, said the task of forecasting, balancing and optimising renewable energy production required “extreme dedication”. Traders needed to monitor short-term factors such as levels of cloud cover, small changes in wind patterns and the effect of ice on wind turbines, he added.
“As the share of renewables in the energy mix continues to increase, the need for real-time balancing of energy markets will only grow bigger,” he said.
Traders also increasingly need to be able to shift their positions at the last minute to account for any surprises in an increasingly complex market.
That has accelerated the adoption of automated trading, according to the study last year by the Netherlands’ ACM.
“The energy transition is a development that drives the use of algorithms even further,” the study said. “The generation of renewable energy is less predictable, as a result of which the need for traders to manage their positions at the last minute increases.”
InCommodities entered the market in 2017, with what Andersen described as a “core belief” that technology and algorithms were “the way forward”.
Energy markets had been shaped by factors including geopolitical tensions, news, policymaking and regulation, advances in technology, demand flexibility, macroeconomic conditions and physical conditions, he pointed out.
“Algorithms are a way to navigate all that data, extract the relevant information and understand how it’s going to impact prices,” Andersen said.
The market has been particularly shaped, meanwhile, by the high profits that traders were able to generate when energy prices rose sharply in 2022 after Russia’s full-scale invasion of Ukraine. InCommodities generated €1.06bn in post-tax profits for 2022, up from €112mn the year before, while Danske Commodities generated €1.47bn on the same measure, up from €303mn in 2021.
While profits have since returned to closer to normal levels, 2022’s bumper returns encouraged the establishment of a series of new competitors. Aros Commodities, Pure Power Trading, Aarhus Energy and Asgard — mostly owned by ambitious young founders — were all established in Aarhus and Aalborg in 2023.
Marc Zimmerlin, a partner at the consultancy Oliver Wyman, said the power trading market had changed as smaller participants built the financial strength to attract staff. Hedge funds and banks were also returning to the market after previously scaling back, he added.
There is growing interest across the range of companies in automated trading. According to Epex, a European power exchange, 70 per cent of volumes traded on the market in 2024 were traded via automated systems, compared with only 44 per cent as recently as 2020.
While some traders rely on “off the shelf” systems bought from others, some market participants have invested heavily in developing proprietary software to give themselves an extra competitive edge.
Adam Perkins, a partner in Oliver Wyman’s energy and natural resources practice, said it was clear from their profit figures that some participants had better models.
The ACM report predicted that automated trading’s share of volumes would continue to increase, as manual trading became less competitive. Algorithms are also growing more sophisticated and, in some cases, can now learn as they go. Such algorithms could improve markets’ liquidity and make price formation more efficient, the study predicted.
However, it also said that the use of algorithms could sometimes inadvertently manipulate the market.
It warned that, when two algorithms engaged in “robot battles” in a market, they could trade so fast that it sent the market “false or misleading signals”.
Acer, a body that co-ordinates the work of Europe’s energy regulators, also expressed concern. It said the rapid speed of algorithmic trading, with orders placed in milliseconds, posed “challenges for surveillance teams at venues and regulators alike”, although it stressed that adequate monitoring systems were in place.
At the office in Aarhus, InCommodities’ McGowan recognised there were some legitimate concerns about the effects of greater algorithmic trading. There could be unexpected outcomes as volumes increased, he said.
“The pace of change over the past few years has been significant,” he said.
New market rules had been introduced to tackle some of the issues, he added, although guidance for companies such as his on what they meant was “virtually non-existent”.
He insisted, however, that his company wanted to get its approach right as it grew.
“We want to pioneer best practices,” he said. “We recognise our responsibilities in the markets.”
Gapping down
In reaction to earnings/guidance:
In reaction to earnings/guidance:
- LPRO -16.7%, GRRR -3.4%
Other news:
- EE -6% (prices offering of 6,956,522 shares of common stock at $26.50 per share)
- AXSM -4% (topline results of Paradigm phase 3 proof-of-concept trial of Solriamfetol in major depressive disorder with and without excessive daytime sleepiness)
- DBVT -2.9% (to delay 10-K filing)
- KLIC -2.9% (winding down Electronics Assembly business)
- TVTX -2.8% (to present new FILSPARI data)
- DHC -2.1% (Closes $140 Million Mortgage Financing Secured by 14 SHOP Properties)
- STM -2% (sign GaN agreement with Innoscience)
- AVA -1.7% (requests price adjustment in Washington; files 2025 Natural Gas Integrated Resource Plan)
- TKR -1.3% (CEO departing; names interim CEO)
- ONON -1.3% (updates leadership structure)
- LPLA -1.1% (prices offering of 4,687,500 shares of common stock at $320.00 per share)
- CMPX -1.1% (Tovecimig meets primary endpoint in the ongoing randomized phase 2/3 study in patients with biliary tract cancer)
Gapping up
In reaction to earnings/guidance:
In reaction to earnings/guidance:
- NUTX +44.6%, PVH +15.7% (also to enter $500 mln ASR), MVST +15.4%, PRGS +7.9% (also names new CEO and COO and files mixed shelf offering), CELC +4.4%
Other news:
- HI +4.7% (completes sale of Majority Stake in Milacron injection molding and extrusion business for $287 mln)
- VCSA +3.9% (Davidson Kempner sent a letter to the Special Committee further enhancing the terms of the prior offers by increasing the cash consideration purchase price to $5.83 per share)
- XRAY +3.7% (files mixed shelf)
- BIGC +3.5% (CTO departing)
- XPEV +3.5% (March deliveries)
- ARDT +2.6% (appoints new COO)
- NIO +2.6% (March deliveries)
- TSSI +1.9% (to delay 10-K filing)
- BTG +1.8% (announces TSX approval for normal course issuer bid)
- AQST +1.4% (topline PK results from its pediatric study and completes the NDA Submission for Anaphylm sublingual film)
- VAL +1.3% ($352 mln contract for drillship)
- INR +1.3% (increases borrowing base under its credit facility)
- APLS +1.1% (FDA has accepted and granted Priority Review designation of the supplemental New Drug Application for EMPAVELI for C3 glomerulopathy and primary immune complex membranoproliferative glomerulonephritis, which are severe and rare kidney diseases)
Research Calls I
-
Upgrades
- AutoZone (AZO) upgraded to Neutral from Sell at Goldman, tgt $3,811
- Enova International (ENVA) upgraded to Buy from Hold at TD Cowen, tgt $17
- First of Long Island (FLIC) upgraded to Outperform from Market Perform at Keefe Bruyette
- First Watch Restaurant (FWRG) upgraded to Buy from Hold at TD Cowen, tgt $22
- Keurig Dr Pepper (KDP) upgraded to Overweight from Equal Weight at Morgan Stanley, tgt $40
- Nayax (NYAX) upgraded to Outperform from Market Perform at William Blair
- Sabre (SABR) upgraded to Market Perform from Underperform at Bernstein, tgt $3.50
- Shake Shack (SHAK) upgraded to Buy from Hold at Loop Capital, tgt $127
- Ulta Beauty (ULTA) upgraded to Buy from Neutral at Goldman, tgt $423
- AutoZone (AZO) upgraded to Neutral from Sell at Goldman, tgt $3,811
-
Downgrades
- American Airlines (AAL) downgraded to Hold from Buy at Jefferies, tgt $12
- Cemig (CIG) downgraded to Underperform from Neutral at BofA Securities
- Delta Air Lines (DAL) downgraded to Hold from Buy at Jefferies, tgt $46
- Genmab (GMAB) downgraded to Underperform from Market Perform at Bernstein
- Genuine Parts (GPC) downgraded to Sell from Neutral at Goldman, tgt $114
- Kroger (KR) downgraded to Sell from Hold at Melius Research, tgt $58
- Rogers Communications (RCI) downgraded to Sector Perform from Outperform at Scotiabank
- Southwest Airlines (LUV) downgraded to Underperform from Hold at Jefferies, tgt $28
- Vipshop (VIPS) downgraded to Neutral from Buy at Citigroup, tgt $18
- W. R. Berkley (WRB) downgraded to Neutral from Buy at BofA Securities, tgt $74
- American Airlines (AAL) downgraded to Hold from Buy at Jefferies, tgt $12
-
Others
- Advanced Drainage (WMS) initiated with a Buy at Deutsche Bank, tgt $134
- Aeva Technologies (AEVA) initiated with a Buy at Canaccord Genuity, tgt $9.50
- Akebia Therapeutics (AKBA) initiated with a Buy at Jefferies, tgt $6
- American Tower (AMT) assumed at Overweight from Equal Weight at Morgan Stanley, tgt $250
- Arbe Robotics (ARBE) initiated with a Buy at Canaccord Genuity, tgt $1.75
- Arcos Dorados (ARCO) initiated with a Buy at BTG Pactual, tgt $10.50
- BlackLine (BL) initiated with a Buy at DA Davidson, tgt $58
- Builders FirstSource (BLDR) initiated with a Buy at Deutsche Bank, tgt $151
- Cameco (CCJ) initiated with an Outperform at Bernstein, tgt $52
- Check Point (CHKP) initiated with an Equal Weight at Stephens, tgt $255
- Core & Main (CNM) initiated with a Buy at Deutsche Bank, tgt $60
- CrowdStrike (CRWD) initiated with an Overweight at Stephens, tgt $450
- Cullen/Frost (CFR) initiated with an Equal Weight at Barclays, tgt $130
- CyberArk (CYBR) initiated with an Overweight at Stephens, tgt $440
- Daimler Truck (DTRUY) initiated with an Underperform at Bernstein
- Ferguson (FERG) initiated with a Hold at Deutsche Bank, tgt $174
- Fortinet (FTNT) initiated at Equal Weight at Stephens, tgt $108
- Fortune Brands Innovations (FBIN) initiated with a Hold at Deutsche Bank, tgt $70
- Griffon (GFF) initiated with a Buy at Deutsche Bank, tgt $91
- Installed Building Products (IBP) initiated with a Hold at Deutsche Bank, tgt $183
- Knot Offshore Partners (KNOP) initiated with a Buy at Fearnley
- Masco (MAS) initiated with a Hold at Deutsche Bank, tgt $71
- Mohawk Industries (MHK) initiated with a Hold at Deutsche Bank, tgt $121
- MongoDB (MDB) initiated with an Outperform at Daiwa, tgt $202
- New Jersey Resources (NJR) initiated with a Hold at Jefferies, tgt $51
- Okta (OKTA) initiated with an Equal Weight at Stephens, tgt $127
- On Holding (ONON) initiated with a Hold at Jefferies, tgt $44
- Owens Corning (OC) initiated with a Buy at Deutsche Bank, tgt $183
- Palantir (PLTR) initiated with a Neutral at Daiwa, tgt $90
- Palo Alto Networks (PANW) initiated with an Equal Weight at Stephens, tgt $205
- Pool Corp (POOL) initiated with a Hold at Deutsche Bank, tgt $344
- Rapid7 (RPD) initiated with an Equal Weight at Stephens, tgt $33
- Rezolve AI (RZLV) initiated with a Buy at H.C. Wainwright, tgt $4
- SBA Communications (SBAC) assumed at Equal Weight from Overweight at Morgan Stanley, tgt $240
- SentinelOne (S) initiated with an Overweight at Stephens, tgt $25
- SiteOne Landscape (SITE) initiated with a Hold at Deutsche Bank, tgt $135
- Snowflake (SNOW) initiated with a Buy at Daiwa, tgt $210
- South State (SSB) initiated with an Overweight at Barclays, tgt $120
- Tenable (TENB) initiated with an Equal Weight at Stephens, tgt $42
- TopBuild (BLD) initiated with a Buy at Deutsche Bank, tgt $356
- Trex Company (TREX) initiated with a Buy at Deutsche Bank, tgt $85
- Volvo (VLVLY) initiated with an Underperform at Bernstein
- Zscaler (ZS) initiated with an Overweight at Stephens, tgt $255
Arm recently sought to acquire Alphawave for AI chip tech, sources say - ReutersSentiment:
Mostly Positive01 Apr 2025 12:00:00 PM
- Alphawave exploring sale after receiving acquisition interest, sources say
- Arm approached Alphawave to buy the chips IP owner, sources say
- Arm has decided not to pursue a deal, sources say
NEW YORK/SAN FRANCISCO/LONDON, April 1 (Reuters) - SoftBank-owned chip tech provider Arm Holdings O9Ty.F recently sought to acquire Alphawave AWE.L, a UK-based supplier of semiconductor intellectual property, to secure a crucial technology that is crucial for building its own artificial-intelligence processors, according to three sources familiar with the matter.Alphawave, which has a market value of 707.1 million pounds ($913.36 million), has been working with its investment bankers to explore a sale after receiving acquisition interest from Arm and other potential acquirers, one of the sources told Reuters.Arm, however, has decided not to pursue a takeover of Alphawave after its initial discussions with the company, two of the sources said.Arm approached Alphawave recently with the aim of acquiring its technology that determines how fast information can get onto and off a chip, which is crucial for AI because chatbots such as ChatGPT and other applications can require thousands of chips strung together at once to operate smoothly.Known as "serdes" - short for serializer-deserializer - the tech is one of the competitive advantages Broadcom AVGO.O boasts, which has helped it win AI chip customers such as Alphabet's GOOGL.O Google and OpenAI.SoftBank and Arm declined to comment. Alphawave declined to comment.Alphawave has a joint venture in China called WiseWave that it operates with the Chinese investment firm Wise Road Capital, which American officials placed on a U.S. blacklist due to national security concerns last year. Arm's ties to China served to complicate its public listing in 2023, Reuters reported at the time.BIG BET ON SERDESUK-headquartered Arm, which is 90% owned by Japan's SoftBank Group 9984.T, does not make chips itself but sells the fundamental building blocks and other intellectual property. It generates revenue by billing companies for a license to use its tech and collects royalty payments for each chip sold.Through a range of tactics Arm has sought to improve its profit margin and expand revenue. These include exploring the idea of designing and selling a chip of its own, signaling a departure from its business of licensing intellectual property to other chip design firms and the possibility of competing directly with Arm's customers.Executives disclosed specifics about the company's fresh approach to its future plans during a December trial in a civil lawsuit over a contract with Qualcomm QCOM.O.The plans disclosed during the trial included internal messages and documents that discussed the introduction of its own chip, though Chief Executive Rene Haas downplayed the comments. He said they reflected long-term strategic spitballing, in which executives often engage.In addition to the details revealed during the trial, Arm has also made a recruiting effort to identify and hire talent capable of launching a chip designed by the company, Reuters reported in February.Arm does not have SerDes technology - short for serializer/deserializer, which converts parallel data to serial data and vice versa for high-speed communication - as advanced as what Alphawave has built. Though little known, this technology serves as the foundation for Broadcom and Marvell Technology's MRVL.O multibillion-dollar bespoke chip businesses, a market projected to grow to as much as $60 billion by 2028, according to a recent Bernstein report from Mark Li.Nvidia has also developed SerDes technology and has indicated it is willing to license it to other companies as part of the custom chip business it launched.Building powerful SerDes technology is crucial to making an AI chip that could stand out from rivals. Building SerDes from scratch requires a specific set of expertise and roughly two years to create, according to chip industry insiders.($1 = 0.7742 pound)(Reporting by Max A. Cherney in San Francisco, Amy-Jo Crowley in London and Milana Vinn in New York; Editing by Matthew Lewis)(( max.cherney@thomsonreuters.com ; 415-404-2697; @chernandburn on Twitter/X))(c) Copyright Thomson Reuters 2025. Click For Restrictions - https://agency.reuters.com/en/copyright.htmlKeywords: ALPHAWAVE-M&A/ (EXCLUSIVE, PIX)nL6N3QE0LH
Copyright © 2025 London Stock Exchange Group plc and its group of companies (LSEG) and/or its affiliates. All rights reserved. The LSEG content received through this service is the intellectual property of LSEG or its third party suppliers. Republication or redistribution of content provided by LSEG is expressly prohibited without the prior written consent of LSEG, except where permitted by the terms of the relevant LSEG service agreement. Neither LSEG nor its third party suppliers shall be liable for any errors, omissions or delays in content, or for any actions taken in reliance thereon. LSEG and its logo are trademarks of LSEG.This e-mail is for the sole use of the intended recipient and contains information that may be privileged and/or confidential. If you are not an intended recipient, please notify the sender by return e-mail and delete this e-mail and any attachments. Certain required legal entity disclosures can be accessed on our website: https://www.thomsonreuters.com/en/resources/disclosures.html
Zealand Pharma
Gubra amylin data impresses, but we remain bullish Petre [rsch.baml.com]
Maintain Rating: BUY
PO:1,000 DKK | Price:491.00 DKK
Equity | 01 April 2025
Key takeaways
· Gubra first multi dose data for its amylin demonstrates impressive 8% weightloss at 6 weeks (directionally stronger vs Petre)
· But key potential differentiator for amylins in obesity is a very clean tolerability profile, with Petre PIb data a high bar
· We continue to see Petre as a potentially best in class amylin with most complete/consistent data and 1.5-2y ahead of Gubra
First Gubra MAD data impressive, but we still like Petre
Today Gubra press released first interim MAD data for its amylin GUBamy (recently licensed to AbbVie, covered by Tim Anderson), demonstrating c8% weightloss on day 43 (c10% placebo adjusted), which looks impressive for the amylin class, and directionally stronger than Zealand's Petre (c5% in a similar trial). However we think the key potential differentiator for amylins in obesity is the improved tolerability profile, with limited colour from the GUBamy data today, and we think Petre's impressive PIb data (see details [rsch.baml.com]) sets a high hurdle for competition. At this stage, we continue to see Zealand's Petre as a potentially best-in-class amylin, with the most complete and consistent dataset across trials and c1.5-2y ahead of Gubra/ABBV. The Cagri data (very similar molecule structure to Petre) also supports our Petre confidence, vs Gubra's less-validated adrenomedullin backbone. As a reminder, Zealand signed a partnering agreement with Roche in March, which we saw as a best-case scenario for Zealand (see our feedback [rsch.baml.com]). Reiterate Buy (see our deep dive [rsch.baml.com]) for a best-in-class amylin now with a clear path to market and a very strong commercialisation partner. Zealand is on our "What's big in SMID" list of top ideas. See overleaf for debates on recent stock weakness
Gubra's first MAD data looks impressive (8% weightloss)
Gubra presented its first interim MAD data for its amylin (dual amylin/calcitonin agonist), demonstrating an impressive 7.77% weightloss on day 43 (vs placebo 1.99% weight gain) - almost 10% placebo adjusted. Release suggested AE's predominantly GI, mild and consistent with the SAD study. In its SAD trial, GUBamy 2.0mg looked fairly tolerable, with 17% GI AEs, and directionally better than the 3.5-6.0mg doses. Beyond data today, Gubra is assessing 3 further doses (we assume the 3.5, 4.75 and 6.0mg doses assessed in its SAD trial) in part B of its MAD over 12 weeks of treatment, with expected completed dosing 4Q25E. Although headline data looks impressive, we caveat: 1) this is headline weightloss data for one dose across 6 patients (and 2 patients in the placebo arm); 2) limited details on tolerability profile, which is our focus as detailed below; 3) surprising GUBamy SAD weightloss curves (maintained weightloss to 43 days, despite an 11 day half-life) are still debated (and 2.0mg saw limited 1 dose efficacy).
Key differentiator for Petre is its tolerability profile
The Petre weight loss (c8-9% at 16-weeks) is sufficient to support its PIII target 15-20% weightloss (and the majority of obese patients who need 10-20% weightloss), in our view, but the key differentiator for us for Petre is its very clean tolerability and safety profile. As a reminder, in its 16-week PIb trial, across the highest two doses, PIb data demonstrated very low tolerability (17% nausea, c4% vomiting) for a PI trial with rapid dose escalations. Additionally it saw: 1) a low portion of moderate events, with c10% of GI events moderate (vs approved GLP1's c20-30%), and often higher for pipeline GLP1's; 2) very clean safety profile, with 2.4/4.8mg doses almost in line placebo, and 9.0mg only slightly worse; 3) ease of titration/ compliance, with only 1 patient discontinuing due to AE's, and only 1 patient delaying dose escalation by 1 week.