Method for synthesizing a radiopharmaceutical

Abstract

The present invention relates to a method for monitoring an automated radiochemical synthesizer during operation, and the radiochemical synthesizer has a plurality of individual activity detectors operably associated therewith. The method includes the steps of recording S10 activity data from each activity detector; accessing S20 historical data from a data storage device; detecting S30 precursors of a decrease in the yield rate in the recorded activity data based on the historical data; predicting S40 the yield rate when synthesizing a tracer using the radiochemical synthesizer based on the detected precursors of a decrease in the yield rate; and initiating S50 an action related to the predicted yield rate level.

Description

Technical Field

[0001] The present invention relates to the field of radiopharmaceuticals during radiopharmaceutical synthesis, such as radiopharmaceuticals used in positron emission tomography (PET) and single photon emission computed tomography (SPECT). Background Art

[0002] PET and SPECT imaging systems are increasingly being used for the detection of diseases and are useful in providing early detection and confirmation of these diseases (e.g., disease states in oncology and neurology). For example, currently, most PET and SPECT tests are related to cancer detection and early Alzheimer's disease detection. These diseases require early diagnosis to allow for timely and effective treatment.

[0003] PET and SPECT imaging systems create images based on the distribution of positron-emitting isotopes and gamma-emitting isotopes in a patient's tissue, respectively. The isotopes are typically administered to the patient by injection of a radiopharmaceutical that includes a probe molecule having: a positron-emitting isotope such as carbon-11, nitrogen-13, oxygen-15, or fluorine-18; or a gamma-radiation-emitting isotope such as technetium-99m. The radiopharmaceutical is metabolized readily and localizes within the body or binds chemically to receptor sites within the body. Once the radiopharmaceutical has localized at the desired site (e.g., chemically bound to a receptor site), a PET or SPECT image is generated.

[0004] Examples of known radiopharmaceuticals include 18 F-FLT ( 18 F]fluorothymidine), 18 F-FDDNP (2-(1-{6-[([2- 18 F]fluoroethyl](methyl)amino]2-naphthyl}ethylidene)malononitrile), 18 F-FHBG (9-[4- 18 F]fluoro-3-(hydroxymethyl)butyl]guanine or [18F]-penciclovir), 18 F-FESP ( 18 F]-fluoroethylpiperone), 18 F-p-MPPF (4-(2-methoxyphenyl)-1-[2-(N-2-pyridyl)-p- 18 p]fluorobenzamido]ethylpiperazine) and 18 F-FDG ( 18 F]-2-deoxy-2-fluoro-D-glucose).

[0005] The radioactive isotopes in radiopharmaceuticals are isotopes that exhibit radioactive decay (e.g., emitting positrons). Such isotopes are commonly referred to as radioisotopes or radionuclides. Exemplary radioisotopes include 18 F, 124 I, 11 C, 13 N, and 15 O, which have half-lives of 110 minutes, 4.2 days, 20 minutes, 10 minutes, and 2 minutes, respectively.

[0006] Because radioisotopes have short half-lives, the synthesis and purification of the corresponding radiopharmaceuticals must be rapid and efficient. Any quality control (QC) assessment of the radiopharmaceuticals must also occur over a short period of time. Preferably, these processes (i.e., synthesis, purification, and QC assessment) should optimally be completed within a time that is below the half-life of the radioisotope in the radiopharmaceutical. Currently, QC assessments (e.g., chemical yield and chemical purity) can be relatively slow, mainly because of the fact that they are performed manually. Accordingly, there is a need for systems, components, and methods for capturing, analyzing, and interpreting data obtained during the synthesis and purification of radiopharmaceuticals to ensure that those syntheses and purifications are performed efficiently so as to produce high-quality radiopharmaceuticals in desired amounts. Based on such analysis, changes can be implemented before, during, or after the synthesis and / or purification of the radiopharmaceutical to correct any deficiencies that occur during the synthesis of the radiopharmaceutical.

[0007] A disadvantage of current solutions is that when a "bad" batch (with a low yield) has already occurred, an absolute steady-state value is used to detect the batch. Accordingly, it is not possible for a user to initiate actions during the process run to prevent a decrease in yield. Summary of the Invention

[0008] The object of the present disclosure is to provide methods and apparatuses that are configured to perform, individually or in any combination, methods and computer programs that seek to mitigate, avoid, or eliminate one or more of the above-mentioned deficiencies and drawbacks in the art.

[0009] This object is achieved by a method of monitoring a radiosynthesizer during operation. The radiosynthesizer has a plurality of individual activity detectors operably associated therewith, and the method includes the steps of: recording activity data from each activity detector; accessing historical data from a data storage device; detecting, based on the historical data, a precursor of a decrease in yield in the recorded activity data; predicting, based on the detected precursor of a decrease in yield, the yield when synthesizing a tracer using the radiosynthesizer; and initiating an action related to the predicted yield level.

[0010] The advantage is that a higher level of product yield can be provided because unexpected decreases in product yield are avoided.

[0011] This object is also achieved by a control system for monitoring an automated radio synthesizer during operation, wherein the radio synthesizer has a plurality of individual activity detectors operatively associated therewith. The control system is configured to: record activity data from each activity detector; detect precursors of a decrease in product yield in the recorded activity data based on historical data accessible from a data storage device; predict the product yield when synthesizing a tracer using the radio synthesizer based on the detected precursors of a decrease in product yield; and recommend actions related to the predicted product yield level.

[0012] Those skilled in the art can obtain additional objects and advantages from the detailed description. Brief Description of the Drawings

[0013] Figure 1 A method for generating and using a PET or SPECT imaging agent and extracting data collection file data according to an exemplary embodiment of the present invention is shown.

[0014] Figure 2 A plot showing data collection file data illustrating the production steps according to an exemplary embodiment is shown.

[0015] Figure 3 A schematic diagram of a system for monitoring the production of radiopharmaceuticals according to an exemplary embodiment is shown.

[0016] Figure 4 A schematic diagram of a sensor array according to an exemplary embodiment is shown.

[0017] Figures 5a and 5b show yield proxies that change with activity loss at tC18 and labelling Δ%, respectively.

[0018] Figures 6a - 6c show precursor events during the labelling reaction.

[0019] Figures 7a and 7b show precursor events during the capture of FTAG on the tC18 cartridge after labelling.

[0020] Figure 8 A plot of the product yield during the process run is shown.

[0021] Figure 9 Shows Figure 8 A portion of the product yield plot in which highlights the first type of anomaly.

[0022] Figure 10Shows a flowchart of the batch browser process.

[0023] Figure 11 Shows a flowchart for monitoring an automated radiosynthesizer during operation. Detailed Description

[0024] Those skilled in the art will readily understand that the embodiments of the invention described herein can have a wide range of uses and applications. Thus, while the invention has been described in detail herein with respect to exemplary embodiments, it should be understood that the embodiments of the present disclosure are illustrative and exemplary and serve to provide a practical disclosure of the exemplary embodiments. The present disclosure is not intended to be construed as limiting the embodiments of the invention or otherwise excluding any other such embodiments, adaptations, variations, modifications, and equivalent arrangements.

[0025] According to an exemplary embodiment of the invention, the following description of different configurations and features is provided. These configurations and features may relate to systems and methods for providing quality control for radiopharmaceuticals and other compounds or preparations containing radioisotopes. While certain names and types of applications or hardware are described, other names and applications or hardware uses are possible, and thus the names provided are only by way of non-limiting example. Additionally, while specific embodiments are described, these specific embodiments are only intended to be exemplary and non-limiting, and it should also be understood that the features and functions of each embodiment can be combined in any combination, which is within the ability of those of ordinary skill in the art.

[0026] The drawings depict various functions and features associated with the exemplary embodiments. While a single illustrative block, subsystem, device, or component is shown, these illustrative blocks, subsystems, devices, or components can be multiplied for various applications or different application environments. Additionally, the blocks, subsystems, devices, or components can also be combined into combined units. Furthermore, while a specific structure or type of block, subsystem, device, or component is shown, this structure is intended to be exemplary and non-limiting, as other structures may be capable of being substituted to perform the described functions.

[0027] Exemplary embodiments of the invention relate to an automated synthesis system for radiopharmaceuticals, which is also referred to and used herein as a "synthesizer" or "radiosynthesizer". The term "automated" means that the synthesizer is programmed such that certain steps are performed in a radiosynthesis operation to produce a tracer. The synthesis system can produce radiopharmaceuticals for use in PET or SPECT scanners. For example, the synthesis system can be the FASTlab from GE Healthcare in Liege, Belgium. TMSystem. The use of the FASTlab system in the examples described herein is intended to be exemplary and non - limiting. It should be understood that the embodiments described herein may be used with a variety of synthesis systems manufactured by companies other than GE Healthcare. It should also be understood that the use of the terms "radiopharmaceutical", "radiotracer", "PET tracer", or "SPECT tracer" herein is intended to be exemplary and non - limiting, and the mention of one term does not exclude the substitution of other terms in the embodiments. Additionally, the term "activity detector" refers to a detection instrument incorporated into an automated synthesizer that detects radioactivity from a gamma source (e.g., a positron - emitting isotope) in its vicinity. Such activity detectors are well - known in the art.

[0028] This disclosure relates to the automated synthesis of radiopharmaceuticals by monitoring a radiopharmaceutical synthesizer during operation and detecting precursors of a decrease in the yield of the final product in the activity data recorded by an activity detector associated with the radiopharmaceutical synthesizer. The precursors of a decrease in the yield are used to predict the yield when synthesizing a tracer in the radiopharmaceutical synthesizer and to suggest actions to increase or maintain the yield based on the predicted yield level. Since performance variations have been observed for automated synthesis devices (e.g., FASTlab synthesizers) at different local manufacturing / synthesis sites, the present invention provides a method for ensuring that the synthesis at each synthesis device is optimized for its production run. Accordingly, this disclosure provides a method for monitoring radiopharmaceutical synthesis using an automated radiopharmaceutical synthesizer and suggesting actions when precursors of a decrease in the yield have been detected, wherein the radiopharmaceutical synthesizer includes one or more activity detectors associated therewith.

[0029] The embodiments disclosed herein contemplate all chemical processes that emit radiation, including but not limited to, for example, nuclear and fluorescence. With regard to nuclear applications, the embodiments include but are not limited to medical isotopes and their corresponding radiation characteristics, such as 18 F, 11 C, 14 C, 99m Tc, 123 I, 125 I, 131 I, 68 Ga, 67 Ga, 15 O, 13 N, 82 Rb, 62 Cu, 32 P, 89 Sr, 153 Sm, 186 Re, 210 Tl, 111 In, or combinations thereof. Preferred isotopes include those for PET, such as 18 F,11 C and 68 Ga.

[0030] Figure 1 depicts a flow chart of a method for synthesizing and using a PET or SPECT imaging agent and extracting data collection file data according to an exemplary embodiment of the present invention. As Figure 1 shown, method 100 can be performed or otherwise executed by one or a combination of various systems, components, and subsystems, including computer-implemented systems. Figure 1 Each box shown represents one or more processes, methods, and / or subroutines implemented in exemplary method 100.

[0031] In block 102, a radioisotope is generated. A cyclotron (e.g., a GE PETtrace 700 cyclotron) for PET radioisotopes or a generator for SPECT radioisotopes (e.g., to produce 99m Tc) is typically used to produce radioisotopes (e.g., 18 F or 11 C). The cyclotron or generator can be located at a manufacturing site or near the scanner. Placing the cyclotron or generator on-site with a PET or SPECT scanner together can minimize the transport time of the radioisotope. It should be understood that although "PET" and "SPECT" are mentioned herein, these examples are exemplary, and the mention of one does not exclude application to the other.

[0032] In block 104, a radiopharmaceutical is synthesized using the radioisotope. The radioisotope is combined with a radioligand using a synthesizer. The result is a radiopharmaceutical. The synthesizer can be manually operated, semi-automatically operated, or fully automated. For example, the GE Healthcare FASTlab system is a fully automated synthesizer. The synthesizer typically operates in a "hot cell" to protect the operator from the radioactivity of the radioisotope. During the synthesis of the radiopharmaceutical, data can be collected during the process. The data corresponds to measurements by radiation detectors or sensors at various points during the synthesis process. The data is collected at various time intervals and can be stored electronically. The data can be output or saved in the form of a data collection file. The synthesizer can employ a cassette that mates with the synthesizer and contains various reagents and other equipment required for radiopharmaceutical synthesis, such as syringe pumps and vials. The cassette can be removable and disposable. The cassette can be configured to support the synthesis of one or more radiopharmaceuticals.

[0033] At block 106, the synthesized radiopharmaceutical is dispensed. A certain dose of the radiopharmaceutical is dispensed into collection vials for patient administration and QC. Samples of a large quantity of the synthesized radiopharmaceutical can be directly dispensed into the QC system and / or a cassette for QC testing. The systems and methods for QC testing are shown in the PCT application numbered US11 / 2011 / 048564 filed on August 22, 2011, the content of which is incorporated herein by reference in its entirety.

[0034] At block 108, quality control checks are performed on the radiopharmaceutical samples. One or more QC checks may be performed. These QC checks can be automated. The QC system can include a cassette having multiple components for performing tests. The cassette can be configured to be inserted into the QC system to perform QC checks. The QC system can be a stand-alone system, or it can be integrated with the synthesizer described above. Radiopharmaceutical doses are dispensed from the synthesizer. Samples can be selected from one or more of the dispensed vials for QC checks. These samples can be sent to the QC system. Alternatively, the QC system can be connected or coupled to the synthesizer such that appropriate samples can be directly output from the synthesizer to the QC system.

[0035] At block 110, doses from the same production batch as the samples that have undergone QC testing are administered to the patient.

[0036] At block 112, a PET or SPECT scan is performed on the patient who has received the dose.

[0037] At block 114, a data collection file is generated from the synthesizer. The file contains data collected during the radiopharmaceutical synthesis. The data collection file can be formatted and contain data as described herein. Alternatively, other formats of the file can be used. For example, the file can be a log file such as generated by the GE Healthcare FASTlab system described above. The use of the terms "data collection file" or "log file" herein is intended to be exemplary and non-limiting, as there are other terms that can be used for such data collection files having data collected during the radiopharmaceutical process. It should be understood that the data collection file can be generated at any point during the synthesis process.

[0038] The data collection file can be generated in a hard copy format and / or can be electronically stored. For example, the data collection file can be printed by an output device (such as a printer) communicatively coupled to the synthesizer. Alternatively, the data collection file can be output or stored in an electronic format. For example, the synthesizer can have an electronic display or be coupled to a computer system for displaying the data collection file in an electronic format. An electronic storage device can be used to electronically store the data collection file, which can be internal or external to the synthesizer. For example, the synthesizer can have a solid state storage device, which can be temporary (such as random access memory) and / or more permanent, such as flash memory or a hard disk type storage device.

[0039] The information from step 116 (indicated by the dashed line) is used in the synthesis of the radiopharmaceutical 104, and for illustration and not limitation, a single batch diagnostic, "yield advisor", or batch browser process is shown as described more fully herein.

[0040] Figure 10 and 11 is a specific / alternative embodiment of step 116 from Figure 1 .

[0041] It should also be understood that the synthesizer can have input devices to allow for user interaction with the system. These input devices can be communicatively coupled to the system. For example, the synthesizer can have a QWERTY or equivalent type keyboard, an alphanumeric keyboard, and / or a pointing input device. A combination of input devices is possible. The synthesizer can be communicatively coupled to a computer network. For example, the synthesizer can be communicatively coupled to a local area network or a similar network. Through such a network connection, the synthesizer can be communicatively coupled to one or more external computers, computer systems, and / or servers. In some embodiments, the synthesizer can be communicatively coupled to the Internet. The synthesizer can be wirelessly connected to a computer network or can be connected through a wired interface. The synthesizer can transmit and receive data through the computer network. For example, the data collection file can be transmitted through the computer network to another computer system or server. The other computer system or server can be remotely located at a geographically separate location from the synthesizer.

[0042] Additionally, the synthesizer can be computer-implemented such that the synthesizer includes one or more computer processors, a power supply, computer memory, and software. As described above, the synthesizer can be communicatively coupled to one or more external computing systems. For example, the synthesizer can be communicatively coupled to an external computer system via a computer network, either wired or wireless or a combination of both. The external computer system can provide commands to cause the synthesizer to operate and to collect and analyze data from the data collection file. This combination of computer hardware and software can enable the synthesizer to operate automatically and perform certain data collection, data analysis, and implementation of factors or corrections derived from the data.

[0043] At block 116, the data collection is analyzed. According to an exemplary embodiment, the data collection file is analyzed as described herein. As part of the analysis, certain factors and information can be collected from the data collection file. Using these factors and information, the radiopharmaceutical process can be changed, modified, and / or adjusted. For example, the data analysis can determine that the process is not operating efficiently because a low yield is indicated. By way of non-limiting example, this can indicate a problem in the reaction vessel. A repair or modification can be implemented. Such a repair or modification can be applied manually by an operator or can be automatically implemented by the synthesizer based on commands issued via the computer system. In some embodiments, the system can be fully automated and does not require external intervention to perform the analysis and implement the correction or modification to the process.

[0044] Figure 2 A diagram depicting how certain information, particularly yield information, can be collected from data in the data collection file according to an exemplary embodiment is shown. Figure 2 a is a representative trace of the radiochemical synthesis of FBA and flucliclatide, and FIG. 200 depicts a plot with an overlay of the components of the radiopharmaceutical synthesis process. The total yield 202 is the sum of the first yield step 204 and the second yield step 206. These yield values can be used to evaluate the performance of the overall process and to identify problem areas in the process. According to an exemplary embodiment, an exemplary or "standard" process with an exemplary yield can be determined for the system. The resulting data collected during the exemplary process, such as measurements from the activity detector, can be plotted. The yield can be determined as Figure 2 shown.

[0045] According to an exemplary embodiment, activity detector No. 1 is located near the tetra-methyl ammonium (QMA) cartridge, activity detector No. 2 is located near the reactor vessel, and activity detector No. 5 is located near the process outlet leading to the syringe or collection vial.

[0046] The resulting plot can form an exemplary "fingerprint" of the system. Subsequent runs using the system can then be compared to this exemplary process. As described above, deviations from the fingerprint can be indicated by plotting the data collection file data. Based on the plot analysis in the comparison, issues regarding the system and its process can be easily identified and subsequently corrected. According to an exemplary embodiment, if a trace is considered the fingerprint of the optimal process, subsequent traces (e.g., from subsequent synthesis runs or from instruments at different sites) can be compared to it. If the fingerprint of the subsequent trace significantly varies (e.g., by more than 2%; more than 5%; more than 10% or more than 15%) in any region (e.g., the regions covered by detectors 1, 2, or 5), the operator (or the synthesizer automatically) can diagnose that the synthesis step is not being carried out correctly. According to an exemplary embodiment, the variations in the first output step 204 and the second output step 206 can be used to identify where in the process problems might occur: in the 18 F]benzaldehyde (FBA) tracer step; in the conjugation step for forming 18 F]fluciclatide; or any purification step involved in the synthesis process.

[0047] Reference Figure 3 , shows a block diagram that provides an overview of a radiopharmaceutical synthesis system. System 10 includes a synthesizer 12 and a controller 14, which has a user interface 16, a processor 18, a program storage unit 20, a storage unit 22, and a communication interface 24. The synthesizer 12 can be any suitable radiopharmaceutical synthesizer, such as, for example, the FASTlab sold by GE Healthcare TM . The synthesizer 12 contains actuators, sensors, and a communication system to perform synthesis runs on cartridges / cassettes / chips and measure the hardware parameters and sensor outputs transmitted to the controller 14. The synthesizer 12 communicates with the controller 14 via a network, which includes but is not limited to a local area network (LAN) 26. Any suitable network arrangement can be implemented to provide communication between the synthesizer 12 and the controller 14, including but not limited to a wide area network or WAN, such as the Internet. The program storage unit 20 stores radiopharmaceutical synthesizer processing programs for synthesizing various radiopharmaceuticals respectively, and other programs if necessary. In addition to the synthesis run data output by the sensors during the synthesis run, the storage unit 22 also stores information such as, but not limited to, reference values / value ranges of the individual sensors in the synthesizer 12. Each radiopharmaceutical synthesized by the synthesizer 12 will have an associated set of reference values / value ranges for the corresponding sensors. These reference values / value ranges can be considered the "reference fingerprint" of a particular radiopharmaceutical synthesis process and / or cartridge. The reference values / value ranges can be programmed into the controller 14 and updated periodically if necessary. For example, as Figure 4As shown, the controller 14 and the synthesizer 12 can also periodically receive reference values / value ranges from a radiopharmaceutical synthesis database system 32 via a network 31 such as the Internet. The system 32 can be maintained locally or globally on a CD, DVD, USB, or some other storage device and processing arrangement. Any suitable communication arrangement can be implemented.

[0048] As previously described, each acquired or measured data can be regarded as an acquired "fingerprint". The acquired fingerprints obtained during the synthesis run can be fed into a failure mode and effects analysis (FMEA), a storage device, or some other similar quality assurance system. For example, the system can be maintained on a local and / or global database, such as potentially multiple donor hospitals, users, and research institutions. In some embodiments, the FMEA can be maintained in the radiopharmaceutical synthesis database system 32. The controller 14 can be located within the synthesizer 12 or at a remote location. In the current embodiment, the synthesizer 12 includes a controller (not shown) to process the commands and data provided by the controller 14 and the information provided by the radiopharmaceutical synthesis database system 32. In some embodiments, the controller 14 can be arranged to initiate a real-time synthesis monitoring process and the controller (not shown) within the synthesizer can run a monitoring program.

[0049] In FIG. 5a, a yield representative of the tC18 activity with respect to the QMA activity is presented (where a 100% yield is the desired result), which varies with the tracer Δ (delta) measured in %. This parameter is defined in FIG. 6b and is related to the yield decrease during the tracer period. According to FIG. 5a, which presents the results of multiple runs from two different locations (Location 1 and Location 2), clearly, there is a correlation between the yield and the Δ% during the tracer period. The outlier indicated by 50 indicates a different failure mode and is a precursor to the yield decrease. The yield is predicted based on the precursor to the yield decrease, and appropriate actions can be recommended to maintain or increase the yield. However, the recommended actions illustrated below are related to the predicted yield level.

[0050] In Figure 5b, the yield representative of tC18 activity related to QMA activity is presented (where a 100% yield is the desired result), which varies with the activity loss measured in % at tC18. This parameter is defined in Figure 7b and is related to the activity loss during capture. According to Figure 5b, which presents the results of multiple runs from two different locations (Location 1 and Location 2), clearly, there is a strong correlation between the yield and the activity loss at tC18. High tC18 losses typically indicate sub-optimal tracers, resulting in low yields, and high tC18 losses are precursors to yield decline. As described above, the yield is predicted based on the precursors to yield decline, and appropriate actions can be recommended to maintain or improve the yield. However, the recommended actions illustrated below are related to predicting the yield level.

[0051] Figures 6a and 6b show precursor events during the tracer reaction. Figure 6a shows a plot similar to Figure 2 the data collection file data, which shows the normalized activity varying with time. As described in connection with Figure 2 above, different activity sensors are used to create the figure. In the region of the figure denoted as 60, the tracer appears, and this section is shown in more detail in Figure 6b. The decline in the normalized activity during the tracer is a measure of the yield decline, denoted as tracer Δ (delta), typically measured in %. As described above, the yield decline during the tracer is an example of a precursor to yield decline.

[0052] According to some embodiments, the step of detecting precursors to yield decline includes detecting anomalies. As described above, the step of detecting anomalies can include measuring the yield decline in a selected region. Anomalies can also be detected by processing historical data from multiple earlier runs to identify behaviors that give early warning signals for the yield. To be able to compare historical data from different runs, the step of processing historical data can further include normalizing data from multiple earlier runs at a specific point. According to some embodiments, the step of detecting anomalies includes fitting a mathematical function to the selected region and evaluating the mathematical function based on its behavior. In this example, region 60 is the tracer reaction, and by evaluating historical data, the mathematical function for this region 60 is selected as:

[0053] y = 1 - Ae^(-λt),

[0054] where y is the yield, A and λ are constants and t is time, and the evaluation is based on the magnitude of λ.

[0055] A large λ corresponds to a "strong curve", which is equal to a "normal" batch with a good yield. Thus, a small λ corresponds to a low yield. Figure 6c is a graph showing the tracer λ varying with the yield representative.

[0056] Based on historical data, many correlations can be identified between the yield and data from the activity detector. Another example is shown in FIGS. 7a and 7b, which illustrate precursor events during the capture of 18F-fluoro-tetraacetylglucosamine (FTAG) (in the region denoted as 70) on a tC18 cartridge after tracer injection. By analyzing historical data from multiple earlier runs in the capture region, behaviors that give early warning signals for the yield may be identified. To be able to compare historical data from different runs, the step of processing historical data may further include normalizing data from multiple earlier runs at a specific point. According to some embodiments, different analysis methods may be used to identify precursor events. These analysis methods include, but are not limited to, comparing time series, identifying Euclidean distances, clustering, etc. Compared with feature-based methods, similarity-based methods require less domain expertise but have lower sensitivity.

[0057] By summing the total "loss" in each of the troughs 71 and 72, a measure of the tC18 activity loss can be established. The result of the historical data analysis is that if the tC18 activity loss is high, this is the result of poor tracer injection and thus a low yield. The yield and the loss are correlated, as shown in FIG. 5b.

[0058] Figure 8 A long-term yield plot during a process run is shown together with an anomaly detection curve. The long-term yield plot includes the raw data and a rolling average - the thick line. An anomaly is a situation that occurs during a process run where the activity reading deviates from what is expected (based on historical data collected during earlier process runs). In this example, one region is highlighted as 75. According to some embodiments, a model is created based on historical data, and the detection of precursors to a yield drop is performed by comparing the recorded activity data with the model.

[0059] Figure 9 Shows Figure 8 A portion of the yield plot in which the first type of anomaly in region 75 is highlighted. The activity readings are used to detect anomalies, which results in the anomaly detection curve, Figure 9 The lower curve in. In this example, the anomaly detected in circle 78 is a precursor event that gives a warning signal and indicates a risk to the yield. The yield can be predicted, and actions related to the yield level can be recommended. According to some embodiments, when the predicted yield level is below a predetermined threshold, actions are taken to maintain the desired output from the radiosynthesizer. According to some embodiments, an automated radiosynthesizer has a production yield of, for example, 75% - 85%, and the predetermined threshold is at least 10% or 15% lower than the yield.

[0060] Examples of recommended actions when a decrease in yield is detected and the yield at the time of synthesizing the tracer is predicted are as follows:

[0061] - When synthesizing the tracer, introduce the material into the automated radiochemical synthesizer, and the recommended actions include adding more material.

[0062] - The recommended actions relate to hardware issues and / or scheduled maintenance, for example, initiate maintenance before the yield drops below an acceptable predetermined level (e.g., 10% or 15% lower than the production yield).

[0063] There are several alternative methods to detect precursors of a decrease in yield. The first option is to use "single batch diagnostics", where deviation behavior is detected. The deviation can be correlated to a trend. This has been illustrated with reference to Figures 6a, 6b, 7a, and 7b for the tracer reaction. Example observations can include: In the case of 18 F-FDG synthesis, the tracer reaction was sub-optimal, showing low FTAG capture. The yield would be sub-optimal.

[0064] This type of problem is most commonly caused by the following reasons:

[0065] 1. 18 Quality parameters of

[0066] 2. FTAG capture problems

[0067] Further actions may be to check the performance against other recent batches. Consider, for example, conveyor line replacement.

[0068] The second option is to use a "Yield advisor", where long-term yield plots with raw data and rolling averages are used. Maintenance events can also be included in the long-term plot. As shown in conjunction with Figure 8 and 9 Automatic anomaly detection can be used to give early warning signals that the yield is at risk.

[0069] Figure 10 Figure 80 shows the flowchart of the batch browser process, which is the third option. The process starts at 80, and a batch is selected at 81. If data from a cluster or similar batches is to be used, as indicated at 82, the process continues to 83. Conversely, if the process does not continue, the process continues to 84, where the selected batch overlaps with historical data from the selected date range or specific batches.

[0070] In 83, the characteristic features of the selected batch are specified to find clusters of similar batches, as shown by 50 in Figure 5a. The process continues from 83 and 84 to 85, where the results are viewed. According to the first option 86, the results are presented as a view trend of the selected batch clusters, and according to the second option 87, the results are presented as a timeline of when batches occur and overlap (e.g., maintenance events).

[0071] In 88, based on the results presented in 85, actions are initiated to prevent a decrease in the yield rate.

[0072] Different considerations regarding connectivity and deployment are envisioned, and this is reflected at different levels of connectivity.

[0073] Local

[0074] In the local version, the different options for detecting precursors of a decrease in the yield rate and the analysis required to predict the yield rate based on the precursors of a decrease in the yield rate must be installed locally as software on the radiochemical synthesizer. The advantage is that there are minimal changes to the site operation. However, the disadvantage is that the update rollout is slow and inflexible.

[0075] Cloud

[0076] In the cloud-based version, manual data is uploaded through a web-based interface, and secure data storage is provided in the cloud. The advantages of a cloud-based implementation are rapid customization, mobile and remote access, data backup, and the possibility of developing new features as the data collection grows. The disadvantage is that it is necessary to prove security and privacy to the system users.

[0077] Figure 11 A flowchart for monitoring an automated radiochemical synthesizer during operation is shown, which has a plurality of individual activity detectors operably associated therewith.

[0078] When generating a tracer in a radiochemical synthesizer, the yield rate is an important parameter. If an unexpected decrease in the yield rate occurs, the system user will not be able to prepare for the event. The purpose of the present disclosure is not only to predict the yield rate itself but also to detect precursors of a decrease in the yield rate. These precursors are associated with the underlying chemical process or the hardware components used in the process.

[0079] The yield rate can be defined as the radioactivity remaining after radiochemical synthesis divided by the radioactivity of the material entering the device. The exact definition may vary for different users. Most users report that fluctuations in the yield rate are the main challenge, although an absolutely high yield rate is also important for some customers (large commercial suppliers). It is important to increase a yield rate that has already decreased, which is achieved by suggesting actions.

[0080] Variations in the yield level (at least for FDG) cause problems. Thus, the claim may be to improve the reliability of the yield.

[0081] In Figure 11 it, the process starts at step S1, and at step S10, activity data is recorded from each activity detector.

[0082] At step S20, historical data is accessed from a data storage device. According to some embodiments, the method further includes selecting S22 historical data to include data from an earlier run on the same radiochemical synthesizer. According to some embodiments, the method further includes selecting S24 historical data to further include data from earlier runs on other radiochemical synthesizers.

[0083] According to some embodiments, the method further includes S26 selecting to arrange the data storage device external to the radiochemical synthesizer, preferably in a cloud-based implementation. According to some embodiments, the method further includes creating a model S28 based on the historical data, and detecting precursors of yield decline by comparing the recorded activity data with the model. According to some embodiments, a local data storage device is arranged within the radiochemical synthesizer, and the method further includes storing S29 the model in the local data storage device.

[0084] At step S30, precursors of yield decline are detected in the recorded activity data based on the historical data. According to some embodiments, the step of detecting precursors of yield decline includes detecting anomalies S32. According to some embodiments, the step of detecting anomalies further includes processing S34 historical data from multiple earlier runs to identify behaviors that give an early warning signal for the yield. According to some embodiments, the step of processing the historical data further includes normalizing data from multiple earlier runs at a specific point.

[0085] According to some embodiments, the step of detecting anomalies includes fitting S36 a mathematical function to the selected region and evaluating the mathematical function based on its behavior. According to some embodiments, the mathematical function is selected as:

[0086] y = 1 - Ae^(-λt),

[0087] where y is the yield, A and λ are constants and t is time, and the evaluation is based on the magnitude of λ.

[0088] According to some embodiments, the step of detecting anomalies further includes measuring S38 the yield decline in the selected region.

[0089] In step S40, the precursor of the detected decrease in the yield rate is used to predict the yield rate when synthesizing a tracer using a radiochemical synthesizer; in step S50, an action related to the predicted yield rate level is initiated. The automated radiochemical synthesizer has a production yield rate that is typically in the range of 75% - 85%, that is, the remaining radioactivity in the output product divided by the radioactivity of the input material. However, this number can vary depending on the type of tracer produced.

[0090] According to some embodiments, the method further includes: when the predicted yield rate level is lower than a predetermined threshold, initiating an action to maintain the desired output S52 from the radiochemical synthesizer. According to some embodiments, the automated radiochemical synthesizer has a yield rate, and the method further includes selecting S54 the predetermined threshold to be at least 10% lower than the production yield rate. According to some embodiments, the method further includes selecting S56 the predetermined threshold to be at least 15% lower than the production yield rate. According to some embodiments, materials are introduced into the automated radiochemical synthesizer when synthesizing a tracer, and the action includes adding more materials. According to some embodiments, the initiated action relates to a hardware problem and / or scheduled maintenance.

[0091] In addition, the monitoring of the automated radiochemical synthesizer can be automatically performed by the on-board computer of the synthesizer. That is, the present invention also contemplates providing a non-transitory computer-readable storage medium having an executable program for performing the steps for monitoring a radiochemical synthesizer such that the execution of the computer-readable program code causes a processor to perform the following steps: recording activity data from each activity detector; detecting, in the recorded activity data, a precursor of a decrease in the yield rate based on historical data accessible from a data storage device; predicting the yield rate when synthesizing a tracer using the radiochemical synthesizer based on the detected precursor of the decrease in the yield rate; and suggesting an action related to the predicted yield rate level in the radiochemical synthesizer.

[0092] The present disclosure includes a computer program for monitoring an automated radiochemical synthesizer during operation, the computer program including instructions that, when executed on at least one processor, cause the at least one processor to perform the method described in conjunction with Figure 11 Furthermore, the present disclosure also includes a computer-readable storage medium carrying a computer program for monitoring an automated radiochemical synthesizer.

[0093] The present disclosure also relates to a controller, as described in conjunction with Figure 3 and Figure 4 The controller is used to monitor an automated radiochemical synthesizer during operation, the radiochemical synthesizer having a plurality of individual activity detectors operably associated therewith, wherein the control system is configured to:

[0094] - Record activity data from each activity detector;

[0095] - Detect precursors of a yield decrease in the recorded activity data based on historical data accessible from a data storage device;

[0096] - Predict the yield when synthesizing a tracer using a radiosynthesizer based on the detected precursors of a yield decrease; and

[0097] - Recommend actions related to the predicted yield level.

[0098] According to some embodiments, the control system is configured to access a data storage device arranged externally.

Claims

1. A method for synthesizing a radiopharmaceutical using an automated radiopharmaceutical synthesizer during operation, the radiopharmaceutical synthesizer having a plurality of individual activity detectors operatively associated therewith, the method comprising the steps of: - Recording activity data from each activity detector (S10); - Accessing historical data from a data storage device (S20); - Detecting precursors of a decrease in the yield based on the historical data in the recorded activity data (S30); wherein the step of detecting precursors of a decrease in the yield includes detecting an anomaly (S32) by processing historical data from a plurality of earlier runs (S34) to identify behaviors that give an early warning signal for the yield, fitting a mathematical function (S36) to the historical data of a selected region and evaluating the mathematical function based on its behavior, wherein the step of processing historical data further includes normalizing data from a plurality of earlier runs at a specific point, wherein the mathematical function is selected as: y = 1 - Ae -λt , where y is the yield, A and λ are constants and t is time, and the evaluation is based on the magnitude of λ, where a small λ indicates a precursor of a decrease in the yield, - Predicting the yield when synthesizing the radiopharmaceutical using the radiopharmaceutical synthesizer based on the detected precursors of a decrease in the yield (S40); - Initiating an action related to the predicted yield level (S50); and - When the predicted yield level is below a predetermined threshold, initiating an action to maintain a desired output from the radiopharmaceutical synthesizer (S52), wherein based on historical data, a precursor event indicates a low radiotracer reaction yield or poor cassette capture.

2. The method according to claim 1, wherein The automated radiopharmaceutical synthesizer has a production yield, and the method further includes selecting the predetermined threshold to be at least 10% lower than the production yield.

3. The method according to claim 2, wherein, The method further includes selecting the predetermined threshold to be at least 15% lower than the production yield.

4. The method according to any one of claims 1-3, wherein Materials are introduced into the automated radiopharmaceutical synthesizer when synthesizing the radiopharmaceutical, and the action includes adding more materials.

5. The method according to any one of claims 1-4, wherein The initiated action relates to hardware problems and / or scheduled maintenance.

6. The method according to any one of claims 1-5, wherein, The method further includes selecting (S22) the historical data to include data from earlier runs on the same radiopharmaceutical synthesizer.

7. The method according to claim 6, wherein, The method further includes selecting (S24) the historical data to further include data from earlier runs on other radiopharmaceutical synthesizers.

8. The method according to any one of claims 1-7, wherein The method further includes selecting (S26) to arrange the data storage device external to the radiopharmaceutical synthesizer.

9. The method according to claim 8, wherein In a cloud-based implementation, the data storage device is arranged external to the radiopharmaceutical synthesizer.

10. The method according to any one of claims 1-9, wherein The method further includes creating a model (S28) based on the historical data, and detecting the precursors of a decrease in the yield by comparing the recorded activity data with the model.

11. The method according to claim 10, wherein, A local data storage device is arranged within the radiopharmaceutical synthesizer, and the method further includes storing (S29) the model in the local data storage device.

12. The method according to claim 1, wherein, The step of detecting an anomaly further includes measuring a decrease in the yield in the selected region (S38).

13. A computer program product for monitoring an automated radiopharmaceutical synthesizer during operation, comprising instructions that, when executed on at least one processor, cause the at least one processor to perform the method according to any one of claims 1 to 12.

14. A computer-readable storage medium carrying the computer program product for monitoring an automated radiopharmaceutical synthesizer according to claim 13.

15. A controller for synthesizing a drug using an automated radiopharmaceutical synthesizer during operation, the radiopharmaceutical synthesizer having a plurality of individual activity detectors operably associated therewith, wherein the controller is configured to: - Record activity data from each activity detector; - Detect precursors of a decrease in the yield rate in the recorded activity data based on historical data accessible from a data storage device; wherein the step of detecting precursors of a decrease in the yield rate includes detecting anomalies by processing historical data from a plurality of earlier runs to identify behaviors that give an early warning signal for the yield rate, fitting a mathematical function to the historical data in a selected region and evaluating the mathematical function based on its behavior, wherein the step of processing historical data further includes normalizing data from a plurality of earlier runs at a specific point, wherein the mathematical function is selected as: y = 1 - Ae -λt , where y is the yield rate, A and λ are constants and t is time, and the evaluation is based on the magnitude of λ, wherein a small λ indicates a precursor of a decrease in the yield rate, - Predict the yield rate when synthesizing a radiopharmaceutical using the radiopharmaceutical synthesizer based on the detected precursors of a decrease in the yield rate; - Suggest actions related to the predicted yield rate level; and - Initiate an action to maintain a desired output from the radiopharmaceutical synthesizer when the predicted yield rate level is below a predetermined threshold, wherein based on historical data, a precursor event indicates a low radiotracer reaction yield or poor cassette capture.

16. The controller according to claim 15, wherein, The controller is configured to access an externally arranged data storage device.

Images