Pennsylvania’s sex-offender registration laws have been the among the most heavily litigated statutes for thirty years. Four versions of Megan’s Law and two versions of SORNA — the Sex Offender Registration and Notification Act — have come and gone, and the appellate courts have struck down pieces of nearly all of them. Defendants have won a lot of these fights.
On August 27 and 28, 2026, the Superior Court published three decisions touching registration. The defendants lost all three.
Each case attacked the registration requirement from a different direction. Michael Shapiro argued that the law’s core assumption about him was factually wrong and that he should get to prove it. Wesley Pollard argued that the statute doesn’t reach him at all, and that if it does, applying it to a 1990 conviction is unconstitutional. Daniel McIntyre didn’t fight the registration at all — by the time his case reached the Superior Court, the lifetime registration and the sexually violent predator label were agreed to, and the fight was over the prison term instead.
Taken together, the three decisions map out how much room is left.
First, the Vocabulary
A few terms come up in all three cases, so it’s worth setting them out plainly.
Tiers. SORNA sorts sex offenses into three tiers based on the offense of conviction, not on anything about the individual. Tier I offenses require 15 years of registration; Tier II, 25 years; Tier III, life. The tier is fixed by the crime.
Sexually violent predator (SVP). This is a separate, individualized designation. The Sexual Offenders Assessment Board evaluates the person and the court decides, after a hearing, whether he meets the statutory criteria. An SVP finding carries lifetime registration plus counseling requirements and community notification. Importantly, a person can be subject to lifetime registration without being an SVP — the tier alone can do it.
Subchapter H and Subchapter I. SORNA II is split into two halves. Subchapter H covers offenses committed on or after December 20, 2012. Subchapter I covers people whose offenses came earlier but whose registration obligations hadn’t yet run out. The two halves have been analyzed separately by the courts, and that distinction matters enormously — as Pollard shows.
Ex post facto. Both the federal and Pennsylvania Constitutions prohibit laws that increase the punishment for a crime after the crime was committed. The prohibition applies only to punishment. If a law is a civil regulatory scheme rather than a criminal penalty, it can be applied retroactively without violating the clause. Nearly every registration challenge of the last decade has turned on which side of that line the statute falls.
Commonwealth v. Shapiro: “Irrebuttable” Means What It Says
Commonwealth v. Shapiro, 2026 PA Super 188 (Pa. Super. filed Aug. 27, 2026) (Stabile, J., joined by Lane, J., and Stevens, P.J.E.).
Shapiro used an instant messaging account to arrange sexual encounters with undercover officers posing as fourteen-year-old girls. In January 2020 he pled guilty to solicitation to commit aggravated indecent assault under 18 Pa.C.S. § 3125(a)(8) and related offenses. He was sentenced to 10 to 23 months of incarceration and a concurrent 48 months of probation.
The Sexual Offenders Assessment Board evaluated him and concluded he was not a sexually violent predator. That did not matter. His offense was a Tier III offense, and Tier III means lifetime registration.
So Shapiro attacked the assumption underneath the tier system. Section 9799.11(a)(4) of SORNA states, as a legislative finding, that “[s]exual offenders pose a high risk of committing additional sexual offenses and protection of the public from this type of offender is of paramount governmental interest.” That finding is the foundation for the whole scheme. Shapiro argued it was unconstitutional as applied to him, because he could prove he personally posed a low risk — and because being publicly labeled dangerous when he isn’t violates his right to reputation under Article I, § 1 of the Pennsylvania Constitution.
The Difference Between “Facial” and “As Applied”
This is the hinge of the case.
A facial challenge says a statute is unconstitutional in all its applications — it can never validly be enforced against anyone. An as-applied challenge is narrower: the statute may be fine generally, but applying it to this person, on these facts is unconstitutional. Losing a facial challenge normally leaves as-applied challenges available. That is the usual sequence, and it is what Shapiro was counting on.
He had a reason to count on it. The Pennsylvania Supreme Court’s decision in Commonwealth v. Torsilieri, 316 A.3d 77 (Pa. 2024) (Torsilieri II), had upheld § 9799.11(4) against what is called an irrebuttable presumption challenge — the argument that a law may not conclusively presume a fact about a person without letting him disprove it. To win that kind of challenge, a party must show three things: an interest protected by due process; a presumption that is not universally true; and a reasonable alternative way to determine the presumed fact.
The Supreme Court in Torsilieri II focused on the second element. And it said something that looked promising for Shapiro: no presumption is true without exception. But it then held that individual variation isn’t the measure. The challenger had to show a scientific consensus rebutting the presumption as to the class of adult sex offenders as a group. Because the appellant’s own experts conceded that adult sexual offenders reoffend at a rate at least three times higher than people convicted of non-sexual offenses, no such consensus existed.
Shapiro’s move was to characterize Torsilieri II as a facial ruling, leaving the as-applied door open. The Superior Court declined.
The reasoning was that Torsilieri II never styled the challenge before it as facial. The Supreme Court evaluated evidence about sex offenders “as a group” and “as a class,” and upheld the presumption because there was no consensus that it was wrong at the class level. Recognizing that individuals within the class differ did not open the door to case-by-case attacks. As the Bucks County trial court put it in a line the Superior Court quoted with approval, “individual dangerousness is not relevant to the statutory scheme.”
The court held that Torsilieri II forecloses any later as-applied challenge whose object is to rebut the presumption — and it does so whether the challenge is labeled substantive due process or procedural due process, because Torsilieri II expressly refused to sort irrebuttable-presumption claims into either box. On the procedural side, the court added a point: an irrebuttable presumption, by definition, forecloses any rebuttal procedure.
The opinion’s own summary: “the word ‘irrebuttable’ means what it says.”
Two footnotes are worth knowing about. Footnote 4 points to Connecticut Department of Public Safety v. Doe, 538 U.S. 1 (2003), where the United States Supreme Court held that Connecticut’s registry did not violate federal due process because it was the fact of conviction, not current dangerousness, that triggered registration — and the conviction came with full process. Footnote 5 notes that this holding stands so long as the irrebuttable presumption doctrine remains viable in Pennsylvania at all, a question the Torsilieri II majority left for another day.
The court also worked through the cases Shapiro cited and explained why none helped. Commonwealth v. Muhammad, 241 A.3d 1149 (Pa. Super. 2020), had found SORNA unconstitutional as applied to a Tier I offender — but that appellant had no criminal history and had not been convicted of a sexual offense at all. Shapiro was convicted of sexual offenses. Commonwealth v. Saunders, 329 A.3d 670 (Pa. Super. 2024), is unpublished and resolved on waiver. And the court noted its analysis matched its own unpublished decision in Commonwealth v. Nunez, 334 A.3d 370 (Pa. Super. Jan. 10, 2025), which had rejected the same argument built on the same kind of individualized expert testimony.
Commonwealth v. Pollard: A 1990 Conviction, and the Chain That Reaches It
Commonwealth v. Pollard, 2026 PA Super 191 (Pa. Super. filed Aug. 28, 2026) (Panella, P.J.E., joined by Nichols, J., and Neuman, J.).
Wesley Pollard was convicted in Luzerne County in 1990 of involuntary deviate sexual intercourse. That was five years before Pennsylvania had any sex-offender registry at all. He was released from prison on July 11, 2003.
In May 2025, the Commonwealth charged him with three counts of failure to register under 18 Pa.C.S. § 4915.2(a)(1), based on an allegation that he had not updated his residence, address, and employment status with the Pennsylvania State Police. He moved to dismiss, arguing that because his conviction predated every registration scheme Pennsylvania has ever had, applying one to him now is retroactive punishment. The Lackawanna County trial court agreed and dismissed all three counts. The Commonwealth appealed.
The Superior Court reversed and reinstated the charges.
The Statutory Question Comes First
The opinion is useful for a reason beyond its holding: Judge Panella walks through the entire history of these statutes in one place — Megan’s Law I (1995) and Commonwealth v. Williams, 733 A.2d 593 (Pa. 1999); Megan’s Law II (2000) and Commonwealth v. Williams, 832 A.2d 962 (Pa. 2003); Megan’s Law III (2004) and Commonwealth v. Neiman, 84 A.3d 603 (Pa. 2013), which struck it down under the single-subject rule; SORNA I (2012) and Commonwealth v. Muniz, 164 A.3d 1189 (Pa. 2017), which held its retroactive application violated the ex post facto clauses; and SORNA II (2018), enacted to fix what Muniz broke. If you have ever had to reconstruct that sequence for a brief, the opinion is worth saving for that section alone.
The court took the statutory question before the constitutional one. Subchapter I applies to two groups. The first — people convicted of a sexually violent offense committed between April 22, 1996 and December 20, 2012 — plainly did not cover Pollard, and both sides agreed on that. The second, § 9799.52(2), covers anyone “required to register with the Pennsylvania State Police under a former sexual offender registration law of this Commonwealth on or after April 22, 1996, but before December 20, 2012, whose period of registration has not yet expired.”
That subsection is what caught him. Pollard was released in 2003, while Megan’s Law II was in force, and he registered under it. Because his underlying offense was involuntary deviate sexual intercourse, Megan’s Law II required him to register for life. So when SORNA II arrived, his registration period had not expired, and § 9799.52(2) applied by its terms.
Note the sequencing that made this work: the hook is not the 1990 conviction. It is the 2003 registration obligation, which the 1990 conviction produced by operation of a statute passed in between.
Then the Constitutional Question
The trial court and Pollard both relied on Muniz and on Commonwealth v. Santana, 266 A.3d 528 (Pa. 2021), for the proposition that retroactive registration is punishment. The Superior Court’s answer was short: both cases interpreted SORNA I. Neither addressed SORNA II, and neither is on point.
The case that is on point is Commonwealth v. Lacombe, 234 A.3d 602 (Pa. 2020), where the Pennsylvania Supreme Court took up whether retroactive application of Subchapter I violates the ex post facto clauses and held it does not. Applying the framework from Kennedy v. Mendoza-Martinez, 372 U.S. 144 (1963), the Lacombe Court found the General Assembly intended a civil regulatory scheme, found three of the five relevant factors weighed toward non-punitive, and held that absent “the clearest proof” that Subchapter I is punitive, a court may not convert what the legislature called a civil remedy into a criminal penalty.
If Subchapter I is not punishment, there is no ex post facto problem. That resolved the case. Order reversed, three charges reinstated. For the proposition that Lacombe reaches someone whose offense predated any registration scheme, the court cited its unpublished decision in Commonwealth v. Strickler, 262 A.3d 508 (Pa. Super. Aug. 17, 2021).
A Jurisdictional Ruling Worth Its Own Note
Pollard filed an application to quash the appeal, relying on the Pennsylvania Supreme Court’s recent decision in Commonwealth v. Sutton, 90 MAP 2024, 104 MAP 2024 (Pa. filed July 21, 2026). Sutton held that 42 Pa.C.S. § 742 does not give the Superior Court jurisdiction over Commonwealth appeals from orders dismissing charges for failure to make out a prima facie case — because in that situation the Commonwealth can simply refile and try again, so the order isn’t final.
The Superior Court denied the application and drew the line in footnote 3. Pollard’s charges were not dismissed for an evidentiary failure. The trial court held the Commonwealth could not charge him with failure to register as a matter of law, which leaves no option to refile. The order was final, and § 742 supplied jurisdiction.
That distinction is going to matter in cases having nothing to do with registration. If you are on the receiving end of a Commonwealth appeal from a dismissal, the question after Sutton and Pollard is whether the dismissal was evidentiary or legal — and the answer determines whether the appellate court can hear it at all.
Commonwealth v. McIntyre: When Registration Isn’t the Fight
Commonwealth v. McIntyre, 2026 PA Super 189 (Pa. Super. filed Aug. 27, 2026) (Stevens, P.J.E., joined by Lazarus, P.J., and Sullivan, J.).
Daniel McIntyre entered an open guilty plea in Montgomery County to thirteen counts of creating child pornography, nineteen counts of possessing it, and one count of corruption of the morals of a minor. The factual basis included photographs he took of a twelve-to-thirteen-year-old child in various states of undress, some involving nitrous oxide and some with the child unconscious, and 3,082 images and 134 videos found on his devices.
He was found to be a sexually violent predator — by agreement of the parties — and ordered to comply with Tier 3 lifetime registration under 42 Pa.C.S. § 9799.15. He was sentenced to an aggregate of 38 to 92 years.
The registration consequence was never contested. That is the reason this case belongs alongside the other two: in a case with the SVP designation and lifetime registration both on the table, neither one was worth litigating. The appeal was about the prison term.
Three holdings came out of it.
No right to an evidentiary hearing on a post-sentence motion. McIntyre argued the sentencing court had to hold a hearing because his post-sentence motion raised arguments not made at sentencing. The court read Pa.R.Crim.P. 720(B)(2)(b) and found it leaves that decision to the trial court’s discretion. His reliance on Commonwealth v. Hubbard, 372 A.2d 687 (Pa. 1977), and Commonwealth v. Dancer, 331 A.2d 435 (Pa. 1975), failed because both were Post Conviction Hearing Act cases about claims that did not appear in the trial record at all. Here, counsel had raised the arguments in a thorough written motion with extensive attachments, and the record showed the trial court reviewed them.
A defective 2119(f) statement can be fixed, but the fix defines the record. McIntyre’s brief omitted the concise statement Pa.R.A.P. 2119(f) requires for a discretionary-sentencing challenge. The Commonwealth objected. He filed a reply brief pointing to his summary of the argument instead, and the court refused to look there — its review is limited to the statement of questions presented and the 2119(f) statement itself. Commonwealth v. Radecki, 180 A.3d 441, 468 (Pa. Super. 2018). He then moved to amend and attached a proper statement, which the court granted, and only then did it reach the merits. The lesson is procedural and cheap: put the 2119(f) statement in the brief.
Parole eligibility in the mid-to-late seventies is not a de facto life sentence. This was the heart of the appeal. A “de facto life sentence” is a term of years so long that it functions as life imprisonment. McIntyre argued 38 to 92 years qualified, since he would not be parole-eligible until age 77 and was statistically unlikely to reach it. He cited Commonwealth v. Dodge, 957 A.2d 1198 (Pa. Super. 2008), where the minimum term ran to age 100½, and Commonwealth v. Austin, 66 A.3d 798 (Pa. Super. 2013), where a 25-year-old received 72 to 192 years and would not be eligible until roughly 97.
The court distinguished both on the numbers and cited its unpublished decision in Commonwealth v. Coleman, 242 A.3d 435 (Pa. Super. 2020), which drew on Commonwealth v. Brooker, 103 A.3d 325, 340 (Pa. Super. 2014): a sentence permitting parole in the defendant’s early eighties is not the equivalent of life. Eligibility in the mid-to-late seventies falls on the same side of that line.
On the merits of the discretionary claim, the court found the sentencing judge had a presentence investigation report, a psychosexual evaluation, and an SVP assessment; heard victim impact testimony and defense expert testimony; and explained on the record why the sentences ran consecutively. Where a sentencing court has the benefit of a presentence report, the appellate courts presume it weighed the defendant’s character and mitigating factors. Commonwealth v. Antidormi, 84 A.3d 736, 761 (Pa. Super. 2014).
One argument the court addressed directly: McIntyre contended that the federal sentence imposed on Ghislaine Maxwell limited what a Montgomery County judge could impose on him. The court called the contention frivolous and moved on.
What this means in practice. If you are challenging a lengthy aggregate sentence as a de facto life term, the number that matters is the age at parole eligibility, and there is now a rough line in the case law: the eighties are not life; Dodge and Austin sit far past that. Know where your client’s number falls before you build the appeal on that theory. And where a presentence report was prepared, expect the presumption that the court considered everything in it — an argument that mitigation was ignored has to point to something the record affirmatively contradicts.