From APHEXDA to cash: how much of BioLineRx’s partnership economics can really turn into clean liquidity

The main article already established that BioLineRx in 2025 no longer looks like a self-commercializing company. It looks like a small junction of rights, royalties, and development programs. This follow-up isolates the question that now matters most for shareholders: how much of APHEXDA's partnership economics can really turn into clean liquidity, and how much gets worn down on the way through equity issuance, lenders, and upstream claims.

That matters because the Gloria and Ayrmid headlines are much larger than the cash that is truly free on the balance sheet. Part of the historical inflow came together with stock issuance, not only with licensing. The recurring money actually visible in 2025 was still very small. And future milestone dollars are not waiting for an empty cash box. Several claims are already standing in line ahead of them.

• Not everything that looks like licensing cash is truly non-dilutive cash. In the Gloria transaction, the 20-F presents combined consideration of $29.6 million, but allocates $12.0 million of that to the share purchase agreement and only $17.6 million to the license agreement. In the Ayrmid and Highbridge transaction, combined consideration was $19.0 million, but $11.2 million was allocated to the registered direct offering and only $7.8 million to the license.
• The recurring pace visible in 2025 is still too small. All 2025 revenue, $1.2 million, came from Ayrmid royalties, against $0.2 million of cost of revenues, while R&D and G&A together were still $11.2 million.
• A future milestone dollar is not a clean dollar. Through December 1, 2027, 10% of future out-licensing milestone payments is earmarked for BlackRock, while motixafortide also carries a 20% Biokine claim on sublicense receipts and a 3.9% IIA rate on sublicense consideration.

The contracts look larger when licensing and equity are bundled together

What matters most in Note 15 and Note 16 is not just the size of the numbers. It is how those numbers were built. In both cases, BioLineRx did not receive only a commercial partner. It also received cash tied to share issuance.

In the Gloria transaction, the clean picture is this: Gloria paid a $15 million upfront payment, while HST and Gloria also bought 170,728 ADSs in a private placement for total proceeds of about $14.6 million. Under the company's own IFRS 15 allocation, out of $29.6 million of combined consideration, $12.0 million was allocated to the equity purchase agreement and $17.6 million to the license agreement. That means a meaningful share of what looks like the "Gloria deal" was effectively equity capital raised through share issuance, not only monetization of the asset itself.

The Ayrmid transaction is even sharper. The company received a $10 million upfront payment from Ayrmid, but at the same time completed a $9.0 million gross registered direct offering to funds associated with Highbridge, a related party to Ayrmid. Here too, the 20-F separates the two layers: out of $19.0 million of combined consideration, $11.2 million was allocated to the registered direct offering and only $7.8 million to the license agreement. In other words, in the transaction that was supposed to signal a shift toward a partnership model, the equity layer was larger than the licensing layer itself.

That chart is not disputing that the cash came in. It did. What it sharpens is the quality of the cash. Cash that comes through licensing and cash that comes through issuing stock are not the same thing for shareholders. The first should be part of APHEXDA's economic conversion. The second arrives together with dilution.

That leads to the first core conclusion of this continuation: the headline size of the two transactions is larger than the non-dilutive liquidity they actually built. Part of the number belongs to commercial partnering. Another part belongs to the capital markets.

What actually arrived in 2025, and what is still mostly on paper

If the 2023 and 2024 headlines are stripped away and the focus moves only to what flowed in 2025, the picture shrinks quickly. BioLineRx's 2025 revenue fell to $1.2 million, and all of it was attributed to Ayrmid royalties from APHEXDA commercialization in stem cell mobilization in the United States. Cost of revenues was $0.2 million, so the recurring stream left about $1.0 million before R&D, G&A, interest, and lease cash.

That is exactly the point that gets lost when investors jump too quickly to the milestone potential. Ayrmid does give BioLineRx rights to up to $87 million of commercial and sales milestones, plus tiered royalties of 18% to 23%. But as of 2025, that layer still does not look like something that funds the company. It looks more like a small starting base that may grow later.

The Gloria side is even more conservative in practical terms. On paper, BioLineRx presents a $15 million upfront payment, up to $49 million of development and regulatory milestones in China and Japan, up to $197 million of sales milestones, and royalties of 10% to 20%. But what actually happened in 2025 does not yet justify reading Gloria as an active liquidity engine. In Note 15, the company explains that in 2024 it recorded an $0.8 million doubtful-accounts provision because of a delayed $2.4 million payment from Gloria, and that the amount was only collected in 2025, which is why the provision was reversed. That is an improvement in collection. It is not proof that the Asia track has already become a new cash stream.

It is also important to separate two different Asia tracks. On the one hand, the 20-F says the SCM bridging study in China under Gloria dosed its first patient in December 2025. On the other hand, the same filing says the randomized PDAC study in China is delayed and that it is unclear when it will be initiated, if at all. So anyone reading Gloria as one broad $49 million development-and-regulatory package is smoothing over a real difference. Part of the path moved forward, but the part that the company itself flags as delayed is also a part that can delay conversion of rights into cash.

This is probably the most important chart in the whole continuation. It shows that for now, APHEXDA has shifted at BioLineRx mainly from a source of economic potential to a small revenue stream that still does not cover the structure. So even if both agreements look impressive, 2025 still does not prove that BioLineRx has become a self-funding royalty company.

Why one milestone dollar is not one clean liquidity dollar

This is where the phrase clean cash needs to be taken seriously. If a Gloria or Ayrmid milestone arrives tomorrow, not all of it will sit freely in the cash box. Part of that money has already been tagged.

The clearest layer sits in Note 10. Through December 1, 2027, 10% of any future milestone payments received under the out-licensing agreements must be used to repay BlackRock principal. That is not a theoretical clause. It is an explicit cash-allocation mechanism.

But it is not the only layer. In Note 14, the company says it owes Biokine 20% of amounts received as consideration in connection with any sublicensing or sale of the licensed technology. In the same note, it says that for motixafortide the IIA rate on sublicense consideration is 3.9%. So on a gross reading of the percentages the company itself discloses, a future milestone dollar that falls inside all three layers is not a clean dollar.

That is an illustrative calculation, not a newly reported line item. It simply combines the three layers the company itself discloses in order to show what the market can easily miss: even if a milestone arrives, a meaningful part of it is already spoken for before it genuinely strengthens shareholder financial flexibility. In BioLineRx's case that is critical, because the real question is not whether value can be created on paper. It is whether enough free cash can be created to extend runway without quickly returning to dilution.

That distinction matters for royalties as well. The 20-F shows $1.2 million of Ayrmid royalties in 2025, but that line already sits against $0.2 million of cost of revenues. Even before loading on all the spending layers that come after it, the first recurring partnership stream is still far from generating true all-in cash flexibility.

Why the ATM is a small bridge, not the answer

If partner cash is still not clean enough and not large enough, the next question is how much room BioLineRx still has in the capital markets. And here too, the 20-F draws a fairly narrow picture.

As of the report date, the company had sold 825,010 ADSs through the ATM program for about $9.6 million of gross proceeds since inception, but under the baby shelf rule it may currently sell only up to another $4.5 million. That is not trivial, but it is not large either. At year-end 2025 the company held $20.9 million of cash, cash equivalents, and short-term bank deposits, but its BlackRock loan still stood at $8.939 million, including $4.479 million of current maturities and another $4.460 million long term.

The simplest comparison is also the sharpest one: the currently available ATM capacity is roughly equal to the current maturities of the BlackRock loan alone, and far below the total BlackRock balance. That means even if BioLineRx were to use the full remaining ATM capacity, it would receive only a relatively small bridge, not a full funding solution.

It is also important to remember what actually happened in 2025 cash flow. The company generated $8.9 million of positive financing cash flow, mainly from the registered direct offering and the ATM, against repayments to BlackRock and lease liabilities. In other words, even after exiting direct commercialization, BioLineRx still relied in 2025 on the capital markets to steady the path. That does not invalidate the partnership thesis. It does mean the thesis is not standing on its own yet.

That is exactly why management still frames the runway so carefully. On the one hand, it says cash and deposits together with other liquidity sources, including royalties and milestones from the Ayrmid and Gloria agreements, should be enough into the first half of 2027. On the other hand, the same filing still carries a going-concern paragraph, and the company continues to say it may explore additional financing sources, including equity offerings. In other words, even in management's own framing, the path from APHEXDA to cash has not yet become a clean and simple one.

Bottom line

The next read on BioLineRx should not start from Gloria's and Ayrmid's maximum headline numbers. It should start from the way those numbers break down. Part of the historical transactions was effectively equity. The recurring stream visible in 2025 is still small. And every future milestone dollar already has layers sitting on top of it that reduce what actually becomes free liquidity.

The current thesis in one line: BioLineRx now holds partnership economics that look rich on paper, but the gap between headline economics and clean liquidity remains wide because the non-dilutive portion is smaller than it seems, the initial royalty pace is low, and part of future milestone cash already belongs in advance to BlackRock, Biokine, and the IIA.

What changed versus the surface reading? Instead of seeing two large licensing deals, investors need to see three separate layers: true asset monetization, capital that came in through equity issuance, and future cash that still depends both on partner execution and on prior claims from other parties.

The strongest counter-thesis is that this reading is still too strict. One can argue that the upfront cash was already received, the company's cost base fell sharply, the BlackRock minimum-cash requirement dropped to $4 million, and only moderate growth in Ayrmid royalties or one meaningful Gloria step is needed to make the liquidity picture much more comfortable. That is a fair argument. But in 2025 there is still not enough numerical proof behind it. For now, the company is still living in the gap between partnership potential and liquidity that is truly available to common shareholders.

Why this matters is straightforward. In a small biotech company, the difference between value that was signed, value that was recognized, and value that is truly free to use is not a footnote. It is the center of the funding thesis. If the market does not separate those three layers, it can easily start treating APHEXDA as if it has already turned BioLineRx into a comfortable royalty story, when in practice it has mainly bought the company more time to prove that this can really happen.