Solid State Characterization of Co-amorphous, Co-crystal and Eutectic systems with Solid-state NMR

Shovik Ray (Indian Institute of Science, India)

LinkedIn: @Shovik Ray; X: @ray_shovik

Abstract: Solid-state nuclear magnetic resonance (SSNMR) spectroscopy is an indispensable tool in pharmaceutical research that provides detailed structural and dynamic insights. This non-invasive technique is particularly critical for identification and quantification of polymorphs forms and tracking of local dynamics. Thus, SSNMR plays a key role at various stages of drug development from preformulation to manufacturing.
In this presentation, I will show the applications of SSNMR in characterization of co-amorphous systems, developed to enhance the solubility and dissolution rates of active pharmaceutical ingredients (APIs). In particular, the molecular interactions leading the co-amorphization of Dasatinib, a tyrosine kinase inhibitor used in chronic myeloid leukaemia therapy, with various co-amorphous systems will be discussed in detail. I will show the use of 1H, 13C and 1H-1H double quantum – single quantum correlation experiments to confirm presence or absence of inter- and intra-molecular hydrogen bonding, which is a key factor in determining formation of co-formers.
I will also discuss recent results on Venetocloax, which is a medication used to treat adults with chronic lymphocytic leukemia (CLL), small lymphocytic lymphoma (SLL) or acute myeloid leukemia (AML). Our studies highlight the capabilities of SSNMR in combination with other characterization techniques to gain pivotal information and optimize pharmaceutical formulations.

  1. Amit Bhattacharya Avatar
    Amit Bhattacharya

    Hi Shovik, nice work! Could you briefly explain how you concluded that intramolecular interactions predominate in the DAS:MAL system based on the DQ-SQ data?

    1. Shovik Ray Avatar
      Shovik Ray

      Hi, Amit, thanks!

      We have executed the DQ-SQ with a DQ excitation time such that, it corresponds to the distance of 1.8-2.3 Å (calculating corresponding H-H dipolar coupling), which is the conventional hydrogen bonding distance. We did get strong intermolecular interactions. But, for DAS-MAL case for the same mixing time I did not get inter-molecular peaks but intra-molecular peaks are present. Therefore, we concluded intra-molecular interaction is predominate in DAS:MAL system, whereas in DAS:SU the inter-molecular interaction is present.

      1. Amit Bhattacharya Avatar
        Amit Bhattacharya

        Thanks Shovik.

  2. Riley Hooper Avatar
    Riley Hooper

    Hi Shovik, nice study. How were the samples prepared – is there any impact of solvent on whether the co-amorphous or co-crystal systems will form? And did you try any other co-former with Ven?

    1. Shovik Ray Avatar
      Shovik Ray

      Hi Riley,

      extremely sorry for the delayed response.

      The samples were prepared by liquid assisted grinding method. Methanol was used as a solvent to co-grind them and to form the co-amorphous system. the impact of solvent we did not check explicitly but yes, if we see the “screening of the co-formers” section in. the poster, with nicotinic acid and phenanthrene, Dasatinib did not form a co-amorphous system, rather a co-crystal. Therefore, I think there is impact of solvent directing co-amorphization/crystallization process. For details please check out (https://doi.org/10.1039/D5CE00064E). We can have more discussions.

      For Ven yes we are studying more co-formers also, and the study is in progress.

  3. Raj Chaklashiya Avatar

    Hi Shovik, interesting study! One question I have is with regards to your deconvolution. I see many different populations in your deconvolution, each with different heights, and was curious about how you performed your fitting procedure (I may have to do something similar for an EPR line in the near future so I could use a few pointers haha!)

    1. Shovik Ray Avatar
      Shovik Ray

      Hi Raj,

      thank you for the question. Let me answer you genuinely.

      The peaks were really broad even after echo-filtering. Therefore, we performed solution NMR to get the chemical shift assignments. Then based on that information and the nature of peaks we performed the deconvolution. The deconvolution was done on topspin (software provided by bruker) with a gaussian/lorentzian model. After initial guesses are given (based on the solution NMR data and their chemical nature), it fitted the spectra. We tried with multiple different initial guesses also, it turned out to be similar end point. So, that’s how we got the deconvolution. If unambiguity of deconvoluted peaks is a question, then to prove that we need the experiments to be done in high sample spinning.

      1. Raj Chaklashiya Avatar

        Gotcha! Thank you for describing how you did that in detail! That is very helpful to know. 🙂

  4. Shovik Ray Avatar
    Shovik Ray

    Hi Raj,

    thank you for the question. Let me answer you genuinely.

    The peaks were really broad even after echo-filtering. Therefore, we performed solution NMR to get the chemical shift assignments. Then based on that information and the nature of peaks we performed the deconvolution. The deconvolution was done on topspin (software provided by bruker) with a gaussian/lorentzian model. After initial guesses are given (based on the solution NMR data and their chemical nature), it fitted the spectra. We tried with multiple different initial guesses also, it turned out to be similar end point. So, that’s how we got the deconvolution. If unambiguity of deconvoluted peaks is a question, then to prove that we need the experiments to be done in high sample spinning.

Leave a Reply

Your email address will not be published. Required fields are marked *