WEBVTT 00:00:00.110 --> 00:00:11.110 It is now my absolute honor, privilege, joy, and opportunity to introduce to all of you Dr. Elizabeth L. Maloney. 00:00:11.110 --> 00:00:19.110 Dr. Maloney is a Minnesota family physician, policy consultant, and medical educator. 00:00:19.110 --> 00:00:29.110 In 2021, she was appointed to the Federal Tick-Borne Disease Working Group, a federal advisory committee on tick-borne illnesses. 00:00:29.110 --> 00:00:37.350 She served on several tick-borne disease working group subcommittees prior to her appointment from 2018 to 2020. 00:00:38.250 --> 00:00:47.170 She has also provided testimony regarding tick-borne disease to several state legislative committees and Canada Health. 00:00:47.170 --> 00:00:58.410 Since 2007, Dr. Maloney has been providing comprehensive, evidence-based, accredited continuing medical education courses. 00:00:58.790 --> 00:01:09.430 Further, she has authored several peer-reviewed papers, including grade-based treatment guidelines on Lyme disease. 00:01:09.430 --> 00:01:18.150 She also speaks to the general public on the need to consistently use effective tick-layer prevention strategies. 00:01:18.630 --> 00:01:21.290 Dr. Maloney, thank you so much for being here. 00:01:23.330 --> 00:01:34.390 Thank you. I want to thank the committee for inviting me to participate in this very wonderful event. 00:01:35.390 --> 00:01:38.730 I have no relationship with commercial interests. 00:01:38.730 --> 00:01:45.750 In my talk today, we'll highlight some of the nuances of the clinically available tests. 00:01:45.850 --> 00:01:48.930 So these are the tests that you're ordering in your clinics. 00:01:49.530 --> 00:01:54.410 So it's going to focus on serology because that's what's available right now. 00:01:54.850 --> 00:02:01.650 Before I do that, though, I'm going to review some basic principles and concepts of laboratory medicine 00:02:01.650 --> 00:02:06.810 because I want to make sure we're all on the same page as we move forward to talk about serology. 00:02:06.810 --> 00:02:15.050 and so testing is not to be put in a vacuum it's part of the diagnostic process and the first part 00:02:15.050 --> 00:02:21.090 of that process is obtaining a history and as you heard from Dr. Alcott there's a lot of questions 00:02:21.090 --> 00:02:26.990 in history to be obtained if you're thinking that someone might have Lyme disease and with a good 00:02:26.990 --> 00:02:33.890 history we can eliminate many potential diagnoses and as we work through our list and 00:02:33.890 --> 00:02:39.910 then we move to clinical exam. An exam will eliminate even more, and so you're at that 00:02:39.910 --> 00:02:45.230 phase where you are able to have a clinical impression. Now, clinical impressions and 00:02:45.230 --> 00:02:52.470 final diagnoses are often the same, but in situations where there's still some uncertainty, 00:02:52.970 --> 00:02:58.270 you can move to use testing to see if you can get to that final diagnosis. 00:02:58.270 --> 00:03:13.410 I want to point out that we have surveyed, I'm not being personally, but there have been surveys of clinicians asking them to rank the relative value of history, exam, and testing, and they did it in that order. 00:03:14.070 --> 00:03:18.710 And I think that's something to bear in mind when those three are not aligned. 00:03:21.280 --> 00:03:26.160 So the goal of diagnostic testing is to help clarify a clinical situation. 00:03:26.160 --> 00:03:32.600 situation. It's going to do that by providing additional information regarding potential 00:03:32.600 --> 00:03:39.360 diagnoses. So diagnostic testing is diagnosis-directed. It is not a fishing expedition. 00:03:40.560 --> 00:03:46.440 So in the case of Lyme disease, might diagnostic testing help when there's a tick bite or an EM 00:03:46.440 --> 00:03:54.340 rash for someone with arthritis who's been exposed to ticks? But how about people in 00:03:54.340 --> 00:04:01.340 endemic areas who have a new second degree or third degree AV block, someone who has cognitive 00:04:01.340 --> 00:04:08.420 problems and peripheral nervous system findings. Or how about the people with PLD, persistent Lyme 00:04:08.420 --> 00:04:14.580 disease? These are the people that have been treated and remain ill. So these would be similar 00:04:14.580 --> 00:04:23.440 to Dr. Ock's would testing help would work out that way. When you're selecting a test, 00:04:23.440 --> 00:04:25.760 is you want to look at various factors. 00:04:25.960 --> 00:04:28.260 So the first is, how is the test performed? 00:04:28.940 --> 00:04:29.680 Is it accurate? 00:04:29.980 --> 00:04:30.560 Is it precise? 00:04:31.460 --> 00:04:32.480 Is it available? 00:04:33.740 --> 00:04:36.200 Is it going to be a lengthy procedure? 00:04:36.360 --> 00:04:37.440 Is it a simple blood draw? 00:04:38.320 --> 00:04:40.920 How costly is it for the patient? 00:04:41.780 --> 00:04:44.480 And then how are you going to use the results? 00:04:44.480 --> 00:04:45.580 Because that matters. 00:04:46.920 --> 00:04:50.300 So you can describe a test by various characteristics, 00:04:50.300 --> 00:04:52.760 and two of them are accuracy and precision. 00:04:53.440 --> 00:04:56.760 So, accuracy gives results that are true. 00:04:57.540 --> 00:05:02.040 So, a positive test means that the patient has the disease, and a negative test means 00:05:02.040 --> 00:05:03.000 that they do not. 00:05:03.780 --> 00:05:06.460 Precision is not interested in disease state. 00:05:06.560 --> 00:05:11.320 They just want to know if the results are about the same when you repeat the test on 00:05:11.320 --> 00:05:11.640 the same. 00:05:12.540 --> 00:05:18.340 Now, the targets here are trying to demonstrate combinations of accuracy and precision. 00:05:19.060 --> 00:05:24.840 So we would want all of our disease patients to be within the target, and we would 00:05:24.840 --> 00:05:27.820 want the people who are disease-negative to be outside. 00:05:28.860 --> 00:05:33.620 They should all cluster because that represents that they're good on repeat 00:05:33.620 --> 00:05:35.120 testing, that they are all the same. 00:05:35.780 --> 00:05:38.220 So in that first, you see that there's great precision. 00:05:38.800 --> 00:05:42.620 The test results have clustered, but they're completely wrong. 00:05:42.620 --> 00:05:45.180 And so you have a precise but inaccurate. 00:05:46.140 --> 00:05:53.220 We flip it around in the middle, so now all the results are correct, but they're scattered, so they're very imprecise. 00:05:53.620 --> 00:05:59.620 And it makes you wonder how long it's going to take to have one of those results fall outside of the area where it should be. 00:06:00.140 --> 00:06:04.920 And then the last, of course, is the ideal test in terms of accuracy and precision. 00:06:06.440 --> 00:06:14.180 But I think clinicians are very interested in two other characteristics, and this is sensitivity and specificity. 00:06:14.180 --> 00:06:18.540 and they actually are the determinants of whether a test is accurate. 00:06:19.660 --> 00:06:24.900 So a sensitive test is able to identify everyone who has the illness. 00:06:25.820 --> 00:06:29.140 A specific test is available to identify those who don't, 00:06:30.080 --> 00:06:32.060 and both are cutoff dependent, 00:06:32.320 --> 00:06:36.760 and you can think of them as opposite ends of a teeter-totter, 00:06:37.360 --> 00:06:40.060 and the fulcrum is the cutoff value. 00:06:40.640 --> 00:06:42.140 So let me show you what I mean. 00:06:42.140 --> 00:06:52.200 If we have a test where results can be anywhere from 0 to 100, and we set the cutoff at 50, it looks like it's favoring specificity over sensitivity. 00:06:52.840 --> 00:06:57.840 But if I shift that fulcrum down to 30, I'm going to have more positive results. 00:06:58.320 --> 00:07:00.280 So I'm going to raise sensitivity. 00:07:00.680 --> 00:07:04.180 But unfortunately, some of those positives are false positives. 00:07:04.640 --> 00:07:06.400 And so specificity goes down. 00:07:06.400 --> 00:07:11.560 If I move that fulcrum in a different direction, I am going to get fewer positives. 00:07:12.140 --> 00:07:14.220 So specificity is gonna go up 00:07:14.220 --> 00:07:16.880 because there'll be fewer false positives, 00:07:16.880 --> 00:07:19.860 but I'm going to increase false negatives 00:07:19.860 --> 00:07:22.340 and so sensitivity goes down. 00:07:22.340 --> 00:07:27.520 So it really does matter where that cutoff is. 00:07:27.520 --> 00:07:31.920 Now, this two by two table shows us disease state 00:07:31.920 --> 00:07:35.000 in columns, positive and then negative. 00:07:35.000 --> 00:07:39.020 And in the rows, you're seeing test results. 00:07:39.020 --> 00:07:41.520 And so the boxes are labeled 00:07:41.520 --> 00:07:44.620 based on whether the test results actually agree 00:07:44.620 --> 00:07:46.600 with the disease state. 00:07:46.600 --> 00:07:49.020 And we can then plug that into equations 00:07:49.020 --> 00:07:54.430 for sensitivity and specificity. 00:07:54.430 --> 00:07:57.070 Sometimes people are interested in predictive value, 00:07:57.070 --> 00:07:59.410 so you can have a positive predictive value 00:07:59.410 --> 00:08:01.250 or a negative predictive value. 00:08:02.250 --> 00:08:04.830 And this is an answer to the question, 00:08:04.830 --> 00:08:07.190 how likely is this positive result, 00:08:08.370 --> 00:08:10.450 something that I can hang my hat on? 00:08:10.450 --> 00:08:13.750 And so people want to know positive result 00:08:13.850 --> 00:08:21.810 Well, a positive predictive value is the proportion of people who have a positive test and actually have the disease. 00:08:22.510 --> 00:08:24.130 And a negative predictive value 00:08:24.130 --> 00:08:25.550 is the proportion of people 00:08:25.550 --> 00:08:27.690 who have a negative test and don't have the 00:08:27.690 --> 00:08:29.730 disease. And so that sounds 00:08:29.730 --> 00:08:31.730 wonderful, but the trouble with predictive 00:08:31.730 --> 00:08:33.510 values, if you look at that equation, 00:08:33.670 --> 00:08:35.570 is you have to know the prevalence 00:08:35.570 --> 00:08:36.550 of disease. 00:08:37.210 --> 00:08:39.330 And unfortunately, that data is not 00:08:39.330 --> 00:08:41.010 available to clinicians. 00:08:41.850 --> 00:08:43.070 And prevalence really matters. 00:08:43.070 --> 00:08:49.610 And here's an example. If we start with a test in an area of 1,000 patients and the prevalence is 00:08:49.610 --> 00:08:55.090 only 10 percent, the table fills out like that. And you can see that the predictive 00:08:55.090 --> 00:09:02.110 value of a positive test is not that great at 64 percent. But if the prevalence of the disease 00:09:02.110 --> 00:09:08.410 increases to 30 percent, now you see a significant increase in the predictive value of 00:09:08.410 --> 00:09:19.120 positive test. In many infections, we can have direct evidence of the organism in question. 00:09:19.880 --> 00:09:26.520 So, direct tests identify the pathogen. Some examples are culture, microscopy, 00:09:26.920 --> 00:09:34.320 PCR, and antigen capture, and we use those in other diseases. There's also indirect evidence 00:09:34.320 --> 00:09:40.560 of infection. So, indirect tests actually are measuring the host response to the pathogen, 00:09:40.560 --> 00:09:43.000 for not measuring the pathogen itself. 00:09:43.680 --> 00:09:47.120 And so many of them are tests of immune activation, 00:09:47.640 --> 00:09:50.660 but some of them look for byproducts of the infection. 00:09:53.080 --> 00:09:57.200 In Lyme disease, it's hard to come by direct evidence. 00:09:58.240 --> 00:10:00.620 Culture and microscopy are not offered, 00:10:00.780 --> 00:10:02.100 and because it's too difficult, 00:10:02.860 --> 00:10:06.820 culture in Borrelia is problematic. 00:10:07.480 --> 00:10:10.120 You need special media with precise growing conditions, 00:10:10.120 --> 00:10:16.320 conditions, and it's a long incubation time. So it's just impractical for labs to offer that. 00:10:16.960 --> 00:10:25.740 A microscopy is a difficult, special stage, very few spirochetes in tissues, and so it's kind of 00:10:25.740 --> 00:10:30.620 like searching for the needle in the haystack. And I think as Dr. Embers mentioned, you know, 00:10:30.660 --> 00:10:36.940 it takes a lot of technical expertise, and I think at one point she said she went through 400 slides 00:10:36.940 --> 00:10:43.420 to find Borrelia. So you can see why that's not awkward. Good news is you can get PCR. 00:10:44.060 --> 00:10:54.300 So these are screenshots from LabCorp, Quest, and Mayo Laboratories. And PCR and positivity is great. 00:10:54.300 --> 00:11:01.900 It's direct evidence of the current infection. You can have multiple specimens that you can try 00:11:01.900 --> 00:11:11.660 to test with PCR so serum spinal fluid synovial fluid and even um tissue biopsies can be tested 00:11:11.660 --> 00:11:18.300 like maybe if you're biopsied and rash um but there's low sensitivity and so you can see the 00:11:18.300 --> 00:11:26.940 various values there um it's highest in synovial fluid but serology is is better than that so PCR 00:11:26.940 --> 00:11:33.280 testing, when positive is great, when negative doesn't really tell you much. PCR tests, as far 00:11:33.280 --> 00:11:42.070 as I know, have not been cleared by the FDA. So they're kind of in-house tests. So what are 00:11:42.070 --> 00:11:47.470 clinicians left with? Well, we're left with serology. And that's the enzyme immunoassays, 00:11:47.710 --> 00:11:54.490 the EIAs, or some of you know them as ELISAs, and then Western blots or immunoblots. And they're 00:11:54.490 --> 00:11:56.430 measuring, there are indirect 00:11:56.430 --> 00:11:57.890 tests and they are measuring 00:11:57.890 --> 00:11:59.870 antibody levels to 00:11:59.870 --> 00:12:02.710 and there 00:12:02.710 --> 00:12:04.210 are many manufacturers 00:12:04.210 --> 00:12:06.310 and dozens of tests on the market 00:12:06.310 --> 00:12:08.310 and the antigens that are used 00:12:08.310 --> 00:12:08.950 in the tests 00:12:08.950 --> 00:12:11.950 these are all FDA 00:12:11.950 --> 00:12:14.250 clear tests and that means 00:12:14.250 --> 00:12:16.190 that when they were trying 00:12:16.190 --> 00:12:17.930 to get on the 00:12:17.930 --> 00:12:19.770 market, they had to demonstrate that they 00:12:19.770 --> 00:12:21.690 were comparable to testing 00:12:21.690 --> 00:12:23.810 that was already out there. They did 00:12:23.810 --> 00:12:29.850 not have to demonstrate clinical validity. There are also something called lab-developed 00:12:29.850 --> 00:12:35.470 tests. These are tests that are designed, manufactured, and used in the lab that created 00:12:35.470 --> 00:12:46.570 them. So this is an even more simplified version of the antibody formation timeline in untreated 00:12:46.570 --> 00:12:53.270 patients. And I really just want to show you that the IgM levels go up first, usually between 00:12:53.270 --> 00:12:57.830 one to two weeks and then peak about four to six weeks and 00:12:57.830 --> 00:13:02.630 then they decline as that IgG begins to rise and we have class 00:13:02.630 --> 00:13:08.790 switching from IgM to IgG. IgG goes up and we think stays up 00:13:08.790 --> 00:13:17.890 indefinitely in the ongoing. Now in 1994 the CDC conducted a 00:13:17.890 --> 00:13:21.330 convened a meeting said you know what we gotta all get on 00:13:21.330 --> 00:13:27.570 the same page about tests because prior to that different institutions were using different tests 00:13:27.570 --> 00:13:33.170 they were interpreting Western blots differently and was causing confusion so a patient could be 00:13:33.730 --> 00:13:41.170 Lyme be a Lyme patient at Yale and not be a Lyme patient at Stone and so they've got people 00:13:41.170 --> 00:13:47.090 together and hashed out what's going to be the test procedures and how we're going to interpret 00:13:47.090 --> 00:13:52.170 Western blots. And that makes sense, especially because, you know, the CDC is in the business 00:13:52.170 --> 00:13:57.170 of epidemiology. You know, you have to understand a disease before you can prevent it 00:13:57.170 --> 00:14:03.090 and treat it. And so they wanted to make sure that they were counting actual caseloads of 00:14:03.090 --> 00:14:12.210 Lyme disease. And so they required a test that was specific. But inherent in their decisions 00:14:12.210 --> 00:14:18.490 were some assumptions about the immune response, specifically that it would be the same regardless 00:14:18.490 --> 00:14:26.790 of patient circumstances. And so that may or may not be true. So here's the two-tier testing 00:14:26.790 --> 00:14:34.770 algorithm. So if you order a Lyme test on your patient, the patient sample will first be tested 00:14:34.770 --> 00:14:42.290 with an EIA. And samples that are negative, we stop. But samples that aren't positive can either 00:14:42.290 --> 00:14:49.670 be tested with immunoblots, which is the standard two-tier testing. Or there's a newer way to 00:14:49.670 --> 00:14:58.250 approach, and that's to test with a second EIA, different than the first, and then that's either 00:14:58.250 --> 00:15:07.090 read as a positive or negative. Now, what I liked in the old diagram that the CDC had was it popped 00:15:07.090 --> 00:15:14.550 if you had a negative EIA, they had the word consider alternative diagnosis, as well you 00:15:14.550 --> 00:15:23.840 should. But this, to my mind, is very different than saying it's not Lyme. In immunoblots, we have 00:15:23.840 --> 00:15:29.880 some special rules on how to interpret them. So if someone's been ill for less than 30 days, 00:15:30.340 --> 00:15:35.800 you do IgM and IgG, and if either one are positive, that's a positive case. 00:15:35.800 --> 00:15:50.480 For people who have been ill for 30 days or more, the rule became you do the IgM and the IgG, but if the IgGs are not positive, it doesn't matter what the IgMs show. 00:15:50.780 --> 00:15:57.600 So positive IgMs after 30 days without IgG positivity, those are false positives. 00:15:58.480 --> 00:16:00.040 And so that's what became accepted. 00:16:02.240 --> 00:16:05.140 Now, two-tier testing is sequential testing. 00:16:05.140 --> 00:16:20.880 And it's really a neat idea. You're going to use a very sensitive test to capture a wide range of patients, and then you'll use a more specific second test to kind of narrow things down to throw out those false positives. 00:16:21.820 --> 00:16:31.920 When you test in this sequence, what happens is that the sensitivity of the two-test sequence is always lower than the sensitivity 00:16:31.920 --> 00:16:38.740 of the second test, and the specificity is always greater, and I'm going to demonstrate it here. 00:16:40.060 --> 00:16:46.240 So another table, we've got 200 patients, and half of them have disease. These are the parameters 00:16:46.240 --> 00:16:54.740 of the first test. Our little two-by-two grid fills out like this. There's 110 people who tested 00:16:54.740 --> 00:17:00.460 positive. 20 of those are false positives, and unfortunately, we missed 10 people, 00:17:00.460 --> 00:17:06.700 Well, we're only going to test the 110, and so if we use a test that is less sensitive 00:17:06.700 --> 00:17:14.300 but more specific, we actually create another 18 false positives, and we have been able 00:17:14.300 --> 00:17:22.780 to successfully winnow down 18 false negatives, and got the false positives down to one. 00:17:22.780 --> 00:17:30.060 So in this sequence, testing was only 72% sensitive, but was 99% specific. 00:17:30.460 --> 00:17:35.220 Here's a cartoon on how that first EIA is done. 00:17:35.220 --> 00:17:38.560 This is with the original one with a whole cell sonicate. 00:17:38.560 --> 00:17:43.160 You're breaking up that bacteria into its various antigens that get embedded in the 00:17:43.160 --> 00:17:45.240 well of the test well. 00:17:45.240 --> 00:17:51.020 And then patient specimens are added, and you can see those little Ys, if they can attach 00:17:51.020 --> 00:17:55.420 to an antigen, those are antibodies, they'll be bonded. 00:17:55.420 --> 00:18:00.740 And we wash that away so the free antibodies go away, and we add a second antibody. 00:18:00.920 --> 00:18:07.840 Now, this is an anti-human antibody, so it's going to bind to the antibodies that remain in the test well. 00:18:07.840 --> 00:18:17.020 And it's called TAG, and TAG means that it's got an enzyme on it that will react with the substrate and cause a color change. 00:18:17.020 --> 00:18:28.620 And so, ultimately, you get a color change that can be read and then ultimately converted to an antibody concentration. 00:18:29.200 --> 00:18:38.010 So, newer tests, newer EIAs are using synthetic or recombinant antigens. 00:18:38.130 --> 00:18:40.470 So, in this case, we're using the C6. 00:18:40.470 --> 00:18:42.750 So, it's the only antigen in the well. 00:18:43.510 --> 00:18:50.290 You can put patient specimens in there, and there might be other antiviral antibodies, but they have nothing to attach to. 00:18:50.730 --> 00:18:52.270 So they'll get washed away. 00:18:52.390 --> 00:18:56.510 And then the rest of the procedure is similar to what I just showed. 00:18:59.870 --> 00:19:01.610 Immunoblots are different. 00:19:01.950 --> 00:19:04.450 So Western blots and Lyme blots are 00:19:04.450 --> 00:19:07.230 different at how they look for antibodies. 00:19:07.770 --> 00:19:11.770 Here, they're actually looking for a specific group of antibodies. 00:19:12.470 --> 00:19:14.490 And this is a very detailed process. 00:19:14.490 --> 00:19:19.910 And it begins by breaking up the Borrelia into different antigens 00:19:19.910 --> 00:19:22.050 and separating them on a gel. 00:19:22.050 --> 00:19:28.410 And so it's a very labor-intensive process, and there's potential bands. 00:19:28.570 --> 00:19:29.850 There are dozens of them. 00:19:30.610 --> 00:19:35.430 And so this becomes very cumbersome to evaluate and actually look at. 00:19:36.270 --> 00:19:43.170 And so there are many performance issues that might add to imprecision of the Western blot, very tech-dependent. 00:19:43.250 --> 00:19:45.390 It's hard to get those antigens to separate. 00:19:45.390 --> 00:19:51.530 And then looking at the intensity of the bands, some of them, on this example, are very dark, 00:19:51.650 --> 00:19:54.010 where others are very, very light. 00:19:55.090 --> 00:19:59.670 And these tests take a long time to get, so you don't get your answer back the next day, 00:19:59.730 --> 00:20:01.410 which is really what you'd like. 00:20:03.570 --> 00:20:09.530 The newer immunoblots are line blots, and what they do is they use recombinant or purified 00:20:09.530 --> 00:20:14.250 antigens, and they can put the exact same amount on each test. 00:20:14.250 --> 00:20:21.310 So you have uniformity there, and they can put the antigens wherever they want. 00:20:21.710 --> 00:20:26.470 They can often do it so that it mimics the alignment of a Western blot. 00:20:27.110 --> 00:20:31.650 And the sample is added, and it's very similar to how the Western blots are performed. 00:20:32.190 --> 00:20:40.530 The big advantage of the immunoblots, it's going to reduce the clutter that we saw in the whole cell sonicate Western blot. 00:20:40.530 --> 00:20:52.250 So back at that 1994 conference, they told you how to do the two-tier sequence, and they also said how to interpret the immunoblots, and 00:20:52.250 --> 00:21:01.710 they wanted it to be very specific. So they took criteria from a study by Engstrom, and here they're only interested in three potential bands. 00:21:01.710 --> 00:21:07.410 You had to have two of the three. And then for the IgG immunoblots, they took 00:21:07.410 --> 00:21:13.730 criteria from a study by Dressler. Now we're looking for potentially 10 different bands that 00:21:13.730 --> 00:21:23.990 we can assess, and you need to have at least five. So some examples of positive IgM, you can see that 00:21:23.990 --> 00:21:30.670 all three samples, one through three, meet the criteria, and that unfortunately four through six 00:21:30.670 --> 00:21:37.670 do not. And indeterminate are the bands that they just couldn't break. It's hard to know sometimes 00:21:37.670 --> 00:21:41.250 if something's really there or if it's just a very faint. 00:21:42.430 --> 00:21:46.710 And so now when we switch to look at the results from an IgG, 00:21:46.890 --> 00:21:51.110 we've got 10 bands to consider, and we want to have five. 00:21:51.210 --> 00:21:53.650 And that's what you see in samples one through six. 00:21:54.250 --> 00:21:56.450 Now, samples four through six are negative, 00:21:56.690 --> 00:21:59.230 but my eye was caught by sample four 00:21:59.230 --> 00:22:02.850 because those four bands are really lighting up. 00:22:02.850 --> 00:22:06.930 The little blue pluses have to do with signal intensity. 00:22:06.930 --> 00:22:12.170 So, the darker the band on the plot or immunoblot, the higher the intensity. 00:22:12.770 --> 00:22:16.390 But intensity is not part of the interpretation criteria. 00:22:18.230 --> 00:22:20.150 So, what do you choose? 00:22:20.390 --> 00:22:23.470 Do you want to do standard two-tier testing or modified? 00:22:23.850 --> 00:22:31.530 Well, modified two-tier testing is thought to be more sensitive in early Lyme disease, so that might be ideal then. 00:22:31.530 --> 00:22:46.890 I, as a clinician, think you get a little bit more information from a standard two-tier testing, especially when the test is negative, because some tests, to my eye, are worth considering or giving a second look. 00:22:47.610 --> 00:22:52.490 This is especially true if you've done your differential and you've tested for the other 00:22:52.490 --> 00:22:55.750 items that you thought might be causing your patient's illness. 00:22:56.270 --> 00:23:01.430 And if you haven't found anything, you might want to come back and say, well, am I being 00:23:01.430 --> 00:23:03.290 too picky about this Western blot? 00:23:03.710 --> 00:23:07.570 Because it turns out there are other Western blot schemes. 00:23:08.270 --> 00:23:14.170 And so they differ on which bands to count, how many you need to have, and whether some 00:23:14.170 --> 00:23:16.330 bands are more important than others. 00:23:18.810 --> 00:23:26.650 Earlier I said that when they selected the two-test sequence and the Western blot 00:23:26.650 --> 00:23:34.540 interpretation that there were some assumptions were made. And so it's my general nature to 00:23:34.540 --> 00:23:41.820 question assumptions, so here we go. The first assumption was that the timing of the IgG 00:23:41.820 --> 00:23:48.220 response was going to be the same across the board, and that's why you had that ignore IgMs 00:23:48.220 --> 00:23:54.480 after one month. But as this quote from Dr. Branda points out, experience is a wonderful 00:23:54.480 --> 00:24:00.460 teacher, and now we know that it can take up to two months for patients to develop an IgG response. 00:24:01.200 --> 00:24:10.680 So, we know the timing of IgG is variable. Another assumption was that there's no impact 00:24:10.680 --> 00:24:16.800 on the patient's immune status when they contracted Lyme disease. We know that immunosuppression 00:24:16.800 --> 00:24:23.400 reduces antibody formation. So, if you have the misfortune of being on chemo for a B-cell cancer 00:24:23.400 --> 00:24:28.040 and you happen to contract Lyme disease, you're not going to mount a response. 00:24:28.740 --> 00:24:35.400 What I couldn't describe from the literature was the impact of corticosteroids. You know, 00:24:35.500 --> 00:24:40.300 most of these trials are small, and so it's great that we have people like Dr. Aucott 00:24:40.300 --> 00:24:45.520 really starting to accumulate patients because you would need a very large patient population 00:24:45.520 --> 00:24:48.600 to then look at the subpopulation of people 00:24:48.600 --> 00:24:51.440 who were on steroids when they got their Lyme disease. 00:24:51.940 --> 00:24:53.300 So that work hasn't been done. 00:24:55.650 --> 00:24:59.850 Another assumption is that the immune response 00:24:59.850 --> 00:25:03.190 would be the same regardless of the clinical presentation. 00:25:04.170 --> 00:25:06.810 And so I'm going to show you two tests that made me, 00:25:07.010 --> 00:25:09.030 or two studies that made me question that. 00:25:09.550 --> 00:25:13.450 So I looked at this paper from Bacon's group, 00:25:13.830 --> 00:25:16.930 and they were looking at how the C6 performs. 00:25:17.150 --> 00:25:24.650 And here you find that when it came to Lyme arthritis, it was 94% sensitive, great test. 00:25:25.190 --> 00:25:29.990 But for late neurologic disease, it was only 73% sensitive. 00:25:32.910 --> 00:25:39.210 And then Dressler's paper, which was the one where we got our 5 of 10 criteria, that 00:25:39.210 --> 00:25:40.970 was a retrospective study. 00:25:41.450 --> 00:25:46.190 So then in the same paper, they reported the results from a prospective study. 00:25:46.190 --> 00:25:53.210 and they were using their own well-characterized patients. And what they did is they took the 00:25:53.210 --> 00:26:00.710 arthritis and neuroborreliosis. And here in this table, they started with 25 with arthritis and 29 00:26:00.710 --> 00:26:09.970 with neuroborreliosis. But it turns out that 24 were four of the 29 neuroborreliosis patients 00:26:09.970 --> 00:26:15.890 or early. So I removed them and I redone the table. And now you can see that the sensitivity 00:26:15.890 --> 00:26:23.450 for arthritis was still great, 96%, but there's a difference. Neuroborreliosis was only 84% 00:26:23.450 --> 00:26:29.850 sensitive. Now, this is theoretical. Well, what if we took the findings from Bacon C6 00:26:29.850 --> 00:26:38.810 and Dressler's IgG side and put them together? We would find that two-tier testing was 90% 00:26:38.810 --> 00:26:46.830 sensitive for Lyme arthritis, but only 61% for late neuroborreliosis. Now, that's in theory, 00:26:46.930 --> 00:26:51.630 but I think it's something that's worth pondering about, and I hope sometime we'll get to the point 00:26:51.630 --> 00:26:56.630 where someone's researching to see if there is actually a difference based on clinical 00:26:56.630 --> 00:27:02.410 presentation. Newer tests have much higher sensitivity, and if you read them, hey, 00:27:02.750 --> 00:27:08.150 we're 100% sensitive for late Lyme disease. That always got me wondering, well, how'd you pull 00:27:08.150 --> 00:27:15.890 that off. Dressler didn't do it. And so I think part of the issue is when they say late Lyme 00:27:15.890 --> 00:27:22.950 disease, they've lumped the two groups together. And the specimens that you're testing against are 00:27:22.950 --> 00:27:30.570 in a performance panel. And to be on a specimen labeled as late in a Performance Panel, you had 00:27:30.570 --> 00:27:37.030 to meet the lab criteria of the CDC definition. So basically, newer tests are finding what had 00:27:37.030 --> 00:27:45.730 already been found. And so what they didn't show us is does this bench sensitivity translate to 00:27:45.730 --> 00:27:51.430 clinical sensitivity? Is it clinically valid? And I also looked at a lot of the papers that 00:27:51.430 --> 00:27:57.250 were describing their results. And when you look at the panels that they were using, 00:27:57.590 --> 00:28:05.290 they were primarily Lyme arthritis. In the CDC's repository that you could test against, 00:28:05.290 --> 00:28:14.750 all of their patients were Lyme arthritis. So here again is that diagram of the antibody response 00:28:14.750 --> 00:28:21.770 in the untreated. And we'd assume that in untreated people that IgG persists indefinitely. 00:28:22.630 --> 00:28:27.390 But we don't know that for sure, because as soon as we find someone, we treat them. 00:28:28.070 --> 00:28:33.790 You know, we can't ethically just say, oh, I'm going to follow you along and see what happens 00:28:33.790 --> 00:28:47.830 So we can turn to animal studies, and Dr. Embers' study in 2012 was interesting in that she had 12 monkeys that were not treated. 00:28:48.010 --> 00:28:57.930 They were infected, but not treated, and wouldn't you know, they all were C6 positive right at five to eight weeks as we would want, and that's what we would expect. 00:28:57.930 --> 00:29:09.630 But then over the course of the next 80 to 90 weeks, in 7 of the 12, or 58%, the C6 reaction became undetectable. 00:29:09.890 --> 00:29:24.550 So it raises the question of, do we have people out there who are untreated with a smoldering infection, but who we can't diagnose if we're reliant exclusively on serology? 00:29:24.550 --> 00:29:31.770 Another assumption is that the results are precise doesn't matter what test you get 00:29:31.770 --> 00:29:37.270 a positive is a positive and a negative is a negative well here we see that this 00:29:37.270 --> 00:29:45.290 group of tests were being tested against the reference labs a reference test and the reference 00:29:45.290 --> 00:29:50.230 test to decide determine whether a test was positive or negative happened to be the Marblot 00:29:50.230 --> 00:30:03.690 But when these researchers used the Marblot against the same samples that the reference lab tested, they were only 85% sensitive for IgG and only 62% for the IgM. 00:30:03.910 --> 00:30:08.550 So replication of results is difficult with Western blots. 00:30:08.550 --> 00:30:17.310 And just to point out that they use the term relative sensitivity and relative specificity as kind of non-standard terminology, 00:30:17.670 --> 00:30:28.170 but because they're doing their comparison against a reference value that may also have missed culture positive cases, this is appropriate terminology. 00:30:28.770 --> 00:30:37.710 So here we have the same lab using the same samples, but different tests. 00:30:37.710 --> 00:30:43.850 And you can see that there's quite a difference between the first and the next two in terms of sensitivity. 00:30:45.390 --> 00:30:55.790 Now, the good news is that those line blots should improve precision because they're eliminating a lot of the technical errors that can occur when you're making a blot. 00:30:59.230 --> 00:31:03.430 People often wonder, what about serology after someone's been treated? 00:31:04.010 --> 00:31:06.830 especially the people who remain ill, 00:31:07.250 --> 00:31:09.990 will it be of any value to test them again? 00:31:10.670 --> 00:31:12.550 And the answer is no, it won't. 00:31:12.890 --> 00:31:15.230 And that's because negative results 00:31:15.230 --> 00:31:17.690 don't indicate that the person is clean. 00:31:18.330 --> 00:31:20.710 We know that antibody values declined 00:31:20.710 --> 00:31:22.750 over time after treatment. 00:31:23.590 --> 00:31:26.270 And positive results don't mean that they're infected 00:31:26.270 --> 00:31:29.530 because that duration of the antibody response 00:31:29.530 --> 00:31:31.690 is highly variable and unpredictable. 00:31:31.690 --> 00:31:39.770 And I know where I live in Minnesota, a lot of doctors think once you're positive, you'll always be positive on testing. 00:31:39.930 --> 00:31:45.770 And that's not true. In most patients, the antibody response will eventually go away. 00:31:48.500 --> 00:31:53.300 So what's the nuts and bolts of testing? What are you going to do in your clinic? 00:31:53.500 --> 00:31:59.080 So if someone walks in with a tick bite, won't that be fun to take that out? 00:31:59.080 --> 00:32:05.720 but should you test them? Well, it's too soon for that antibody, so don't make the mistake of 00:32:05.720 --> 00:32:12.340 testing, because if you test and you happen to get a positive result, that's really evidence of 00:32:12.340 --> 00:32:18.920 higher exposure, and I don't think it's coming from that particular bite. And similarly, in people 00:32:18.920 --> 00:32:23.840 with post-treatment Lyme disease or persistent Lyme diseases, whatever term you're going to use, 00:32:23.840 --> 00:32:32.040 We know that antibiotics can change the immune response, and so whatever you get is not going to be a reliable indicator. 00:32:32.240 --> 00:32:36.120 So testing really won't help me, at least not serology. 00:32:37.540 --> 00:32:46.860 Now, Lyme arthritis is interesting, and that's because more and more, it seems that people are showing up with arthritis without a preceding erythema migrans. 00:32:46.860 --> 00:32:54.900 and they also there's not seasonality to Lyme arthritis it can occur within just a couple months 00:32:54.900 --> 00:33:02.020 of being bitten by a tick or might emerge years later and so that makes it a trickier diagnosis 00:33:02.020 --> 00:33:07.800 based on clinical grounds but you want to be able to separate out someone who has Lyme 00:33:07.800 --> 00:33:13.820 arthritis versus someone who is a different bacteria causing their septic joint and so here 00:33:13.820 --> 00:33:20.860 testing is quite valuable because it has high specificity and high sensitivity. So test away 00:33:20.860 --> 00:33:29.160 if you're worried about Lyme arthritis. In these situations, Lyme disease testing can be useful 00:33:29.160 --> 00:33:35.380 when it's positive because especially in an endemic state like Delaware, if someone comes in 00:33:35.380 --> 00:33:44.740 with AV block who's had exposure or a facial nerve palsy or a smoldering type of meningitis, 00:33:44.740 --> 00:33:47.980 or even those adults who get the summertime flu, 00:33:48.580 --> 00:33:51.980 if they test positive, you can trust that that's a true positive. 00:33:52.480 --> 00:33:55.560 Same with people who present with late neuroborreliosis. 00:33:56.460 --> 00:33:59.120 Remember that in very early cases, 00:33:59.300 --> 00:34:03.880 it might be that the modified two-tier testing would be more sensitive. 00:34:04.920 --> 00:34:07.460 The trick here is if you get negative results, 00:34:07.920 --> 00:34:10.780 don't throw away the idea that this could be Lyme 00:34:10.780 --> 00:34:14.300 because you might find out once you work through your differential 00:34:14.300 --> 00:34:17.460 that Lyme makes more sense than anything else. 00:34:19.420 --> 00:34:25.140 Do not test if you see someone with an erythema migrans rash. 00:34:25.660 --> 00:34:27.920 It's not the pathogen of Lyme, 00:34:28.000 --> 00:34:31.160 but because we do know that STARI and other things 00:34:31.160 --> 00:34:33.100 can cause a similar appearing rash. 00:34:33.420 --> 00:34:36.320 But it is a hallmark lesion of early Lyme disease, 00:34:36.840 --> 00:34:41.540 and you don't want to delay treatment and be confused by test results 00:34:41.540 --> 00:34:49.300 because, as you see from Dr. Horn's study, she had only 29% of the people that had definitely 00:34:49.300 --> 00:34:56.960 had an EM rash were two-tier positive. And so, use your clinical judgment, look at the clinical 00:34:56.960 --> 00:35:06.320 evidence, and then make your diagnosis. One of the tricky ones are people who fail to do Lyme 00:35:06.320 --> 00:35:11.460 prevention, and they've got Lyme not only once, but perhaps twice. And, so, you're trying to 00:35:11.460 --> 00:35:18.320 discern if someone comes in with an EM rash with a history of Lyme disease. Is this a new case and 00:35:18.320 --> 00:35:24.680 they've been reinfected or are they simply having a relapse of their disease? Because both can occur. 00:35:26.160 --> 00:35:32.620 Testing here is tricky. If you want to look at the details, one of the references is to Dr. 00:35:32.620 --> 00:35:38.460 Branda's 2021 paper. He does a nice section on 00:35:38.460 --> 00:35:42.940 reinfection versus relapse. So this is when you're going to rely more on 00:35:42.940 --> 00:35:47.260 clinical information. Did the patient have a new exposure history? 00:35:47.260 --> 00:35:52.780 Did their EM rash, is it in the 00:35:52.780 --> 00:35:56.380 a different location than their previous EM rash? 00:35:56.380 --> 00:35:59.340 Do their symptoms, are they different in this 00:35:59.340 --> 00:36:02.100 in illness than in their previous. 00:36:02.100 --> 00:36:04.720 If the EM rash is at the same place, 00:36:04.720 --> 00:36:07.540 then that speaks more likely to a relapse 00:36:07.540 --> 00:36:11.520 because it would be, I hope, unusual for an EM 00:36:11.520 --> 00:36:17.120 to be exactly in the same location on reinfection. 00:36:17.120 --> 00:36:21.000 So to summarize, current testing, 00:36:21.000 --> 00:36:24.460 serologic testing for Lyme disease is limited 00:36:24.460 --> 00:36:28.400 because it's insensitive in certain situations. 00:36:28.400 --> 00:36:42.320 However, the specificity is wonderful, and so someone in Delaware with a good clinical history suggestive of Lyme disease and two-tier testing that's positive certainly supports the diagnosis. 00:36:42.640 --> 00:36:53.020 You always have to consider lab results in the context of an individual's risk factors, history, and exam findings. 00:36:53.020 --> 00:37:00.480 and until we have good direct tests of infection you're going to have to use your clinical judgment 00:37:00.480 --> 00:37:06.600 as you diagnose patients with Lyme disease. Thank you and I look forward to any questions. 00:37:06.960 --> 00:37:18.460 I can talk. Okay. 00:37:20.540 --> 00:37:28.120 So just a little bit of a I guess a technical question about the Western blot. I remember 00:37:28.120 --> 00:37:32.880 one of your slides you had like the numbers for a patient had like 39 kilodaltons there was 00:37:32.880 --> 00:37:41.220 a band or whatever so like i guess so you're my teacher if i see a band at 39 kilodaltons and that 00:37:41.220 --> 00:37:44.940 band is there that's from Western blot that had a secondary antibody and stuck to a first 00:37:44.940 --> 00:37:48.420 antibody that was created against the protein that's the rolling protein how is that negative 00:37:48.420 --> 00:37:53.720 just in a generic like because i like it's the part you're saying about there's four bands but 00:37:53.720 --> 00:37:59.540 how can there be a band and still be negative if that does that make sense i understand what 00:37:59.540 --> 00:38:05.380 you're asking because you're you're saying hey this is an important band 39 and it's showing up 00:38:05.380 --> 00:38:13.120 um why would i not just on the basis of that say that my patient has Lyme disease and it you know 00:38:13.120 --> 00:38:20.380 some of these antibodies antigens cross-react with other uh infectious agents and so that's why 00:38:20.380 --> 00:38:26.380 they don't hang their hat on a single band and that's they really want to be specific so that'S 00:38:26.380 --> 00:38:34.540 they came up with the five or ten criteria but other um schemes for interpreting Western blots 00:38:34.540 --> 00:38:41.740 do say you know you these two bands are important um and that they're sufficient for a diagnosis 00:38:41.740 --> 00:38:49.500 but what was decided on at the 1994 conference was you had to have this type of band pattern 00:38:49.500 --> 00:38:54.890 oh but that makes sense thank you i don't need it either sorry 00:38:54.890 --> 00:39:02.410 if you've got five of those ten that are positive and you have the 00:39:02.410 --> 00:39:07.770 Western blot that's also interpreted as positive is there any way that that could be a false positive 00:39:11.370 --> 00:39:18.730 sure so the question is if you got a positive uh Western blot so uh the first two tests 00:39:18.730 --> 00:39:25.530 first tier test was positive and now you have the positive uh Western blot um is there any way that 00:39:25.530 --> 00:39:31.210 that could be a false positive well yes so you know if you do the numbers there's like a less 00:39:31.210 --> 00:39:38.730 than one percent chance uh that that's a false positive uh but that means there's a 99 chance 00:39:38.730 --> 00:39:48.070 that it didn't so especially in endemic states you do want to trust the positive results yes 00:39:48.070 --> 00:39:53.110 I have a question about Lyme arthritis. How do you determine if the testing is negative 00:39:53.110 --> 00:39:57.990 if the patient wasn't treated and they have arthritis syndrome but she's not in the toilet? 00:39:59.830 --> 00:40:03.190 Oh darn. So the question was how do you 00:40:03.190 --> 00:40:07.750 determine osteoarthritis from Lyme arthritis if the test is negative? 00:40:08.710 --> 00:40:11.430 You know the testing was so good for Lyme 00:40:11.430 --> 00:40:17.190 arthritis that I would not think that you're getting a false negative. 00:40:17.190 --> 00:40:20.870 I think that you have to trust that test that they don't have Lyme arthritis. 00:40:21.310 --> 00:40:26.510 It does get trickier as people age because we do develop osteoarthritis. 00:40:28.110 --> 00:40:34.070 When I was practicing, I'm not practicing now, but when young people came in with their Lyme 00:40:34.070 --> 00:40:36.930 arthritis, they were always positive. 00:40:37.270 --> 00:40:43.610 And the other interesting thing about Lyme arthritis versus osteo, the joint can really blow up. 00:40:44.270 --> 00:40:45.850 And yet they kind of hobble in. 00:40:45.850 --> 00:40:53.310 Whereas if you have septic arthritis from another cause, they're not hobbling and they're coming in in a wheelchair. 00:40:53.850 --> 00:41:03.130 And so there usually is clinical difference between osteoarthritis, Lyme arthritis, and then at the extreme end is septic arthritis. 00:41:08.270 --> 00:41:08.870 All right. 00:41:09.130 --> 00:41:10.670 Take the mic to the person in the short. 00:41:10.870 --> 00:41:11.250 Thank you. 00:41:13.420 --> 00:41:21.200 Something that I've read on false positives was having previous diagnoses of mononucleosis. 00:41:21.640 --> 00:41:24.940 Is that something that you could maybe expand on just a little bit? 00:41:25.240 --> 00:41:25.600 Well, sure. 00:41:25.820 --> 00:41:30.530 So there are cross-reacting antibodies. 00:41:30.830 --> 00:41:40.050 So if we think of bacteria, they can have similar antigens on their surface, or in this 00:41:40.050 --> 00:41:41.470 case with mono or virus. 00:41:41.830 --> 00:41:47.610 And so you can get cross-reacting antibodies, and so that is another reason why they didn't 00:41:47.610 --> 00:41:54.110 hang their hat on a single antibody, but actually had to say, you know, you have to have more than 00:41:54.110 --> 00:42:02.930 one. But what you're getting at is an interesting point. And so specificity is great if you're an 00:42:02.930 --> 00:42:08.730 epidemiologist or a clinical researcher, because you need to make sure that you're not counting 00:42:08.730 --> 00:42:15.450 or studying people who don't have your disease. But a clinician, you want to have sensitive tests 00:42:15.450 --> 00:42:20.490 because you're trying to find anyone who might benefit from your therapeutic intervention. 00:42:21.170 --> 00:42:24.490 And so there's kind of a dynamic tension between the two. 00:42:25.470 --> 00:42:31.710 I'm also very cognizant that we don't want to give people a label of Lyme disease if they have something else. 00:42:32.150 --> 00:42:36.630 When I speak to the public, I always say the goal is not to be diagnosed with Lyme 00:42:36.630 --> 00:42:41.390 disease. It's to be diagnosed correctly. And so that's that's the basis. 00:42:41.390 --> 00:42:51.430 but yes um obviously with a lot of research that you're referencing also is being referenced by a 00:42:51.430 --> 00:42:56.310 lot of the other institutions that are teaching this thing um a lot stuff has happened since 00:42:56.310 --> 00:43:04.130 about 20 but the CDC is basing a lot if there are results on 799 is there any way that they're 00:43:04.130 --> 00:43:08.710 going to be updating things soon because it sounds like things have kind of gotten out of date and 00:43:08.710 --> 00:43:13.210 And with presentations like this, when it endures you soon, it looks like there's a 00:43:13.210 --> 00:43:18.250 lot of development in the Lyme world and in the CDC going to be hopefully updating things 00:43:18.250 --> 00:43:24.510 a little soon because the five-band testing protocol limitation is a problem. 00:43:25.490 --> 00:43:27.050 Do I need to repeat that? 00:43:27.290 --> 00:43:33.270 It's basically, is the CDC gonna be kind of modernizing their criteria? 00:43:34.010 --> 00:43:38.270 And I can't speak for the CDC, but I am gonna go to a round table that they're offering 00:43:38.270 --> 00:43:41.310 next week on their vector borne disease policy. 00:43:43.570 --> 00:43:49.570 You know, when I read these papers from the 90s and moving forward, it's always this promise 00:43:49.570 --> 00:43:52.510 of the test, the good test is right around the corner. 00:43:53.110 --> 00:43:57.210 And we're really needing testing to be direct testing, right? 00:43:57.610 --> 00:44:02.670 And so I think that we have other people in the room that can speak to the future of testing. 00:44:03.630 --> 00:44:06.170 Sadly, we're stuck right now with what we've got. 00:44:06.170 --> 00:44:13.010 But there again, that first thing I said, how testing doesn't occur in a vacuum. 00:44:13.770 --> 00:44:20.690 And so, yeah, sometimes you're going to move forward with treatment in someone who doesn't meet that two-tier standard. 00:44:21.350 --> 00:44:28.290 But what I would suggest to you is someone who's done a lot of peer review in general, whenever you're 00:44:28.290 --> 00:44:32.710 going off in a different direction, you just have to document your thinking quite well 00:44:32.710 --> 00:44:38.350 and so that people can follow how you arrive at the decision to test or treat 00:44:38.350 --> 00:44:42.490 when things aren't quite what they are told to do. 00:44:42.490 --> 00:44:58.310 My question is, in human medicine, do you have a vaccine approved for Lyme disease in particular? 00:44:58.810 --> 00:45:02.450 So the question is do we have a vaccine for Lyme disease? 00:45:02.450 --> 00:45:09.110 Not at the moment, but I've got my fingers crossed that there's some vaccines in development. 00:45:12.310 --> 00:45:19.090 I'm just curious, because in veterinary medicine, we have the Lyme vaccine for a long time already. 00:45:19.970 --> 00:45:23.410 Right. I know that. And there used to be a human vaccine. 00:45:23.910 --> 00:45:27.450 It was withdrawn from the market in 2002. 00:45:28.170 --> 00:45:33.310 And depending on who you listen to, there's various reasons for it being withdrawn. 00:45:33.310 --> 00:45:38.350 I think that we're careful about creating a vaccine. 00:45:38.550 --> 00:45:44.590 We don't want to cause injuries in a vaccine, so there is one being tested now. 00:45:44.670 --> 00:45:47.870 I think they might be in stage three clinical trials. 00:45:48.450 --> 00:45:56.490 So I really want a vaccine because it's really hard to get people to practice prevention, 00:45:57.290 --> 00:46:01.410 and so I think a vaccine is going to be very useful. 00:46:01.410 --> 00:46:07.800 So hopefully it'll be soon. 00:46:07.800 --> 00:46:08.640 All right. 00:46:08.640 --> 00:46:09.860 Well, it looks like I'm done. 00:46:09.860 --> 00:46:11.440 I want to thank you for your attention. 00:46:11.440 --> 00:46:12.280 Have a lovely.